| *** sdake_ has quit IRC | 00:00 | |
| *** BadCub is now known as BadCub_Hiding | 00:01 | |
| *** tjones1 has left #openstack-nova | 00:02 | |
| *** hdd has joined #openstack-nova | 00:04 | |
| mikal | . | 00:05 |
|---|---|---|
| *** yingjun has quit IRC | 00:07 | |
| *** mriedem has joined #openstack-nova | 00:07 | |
| jogo | mikal: .. | 00:08 |
| *** promulo__ has joined #openstack-nova | 00:08 | |
| jogo | mikal zz_johnthetubagu : btw I think its time to start a liberty priorities etherpad | 00:08 |
| dansmith | we did that at the last summit, right? | 00:09 |
| mikal | jogo: sure, wanna do that thing? | 00:09 |
| mikal | We can gather proposals now | 00:09 |
| jogo | dansmith: we started the etherpad well before hand | 00:09 |
| mikal | And then pick winners at the summit | 00:09 |
| dansmith | okay | 00:09 |
| dansmith | yeah, I guess that's right | 00:09 |
| *** matrohon has quit IRC | 00:10 | |
| *** eliqiao has joined #openstack-nova | 00:10 | |
| *** park_heijlong has joined #openstack-nova | 00:10 | |
| *** mspreitz has quit IRC | 00:10 | |
| jogo | http://lists.openstack.org/pipermail/openstack-dev/2014-October/047914.html | 00:10 |
| jogo | mikal: posting that now | 00:10 |
| mikal | jogo: ta | 00:11 |
| *** promulo has quit IRC | 00:12 | |
| *** igordcard_ has quit IRC | 00:15 | |
| *** annegentle has joined #openstack-nova | 00:16 | |
| jogo | https://etherpad.openstack.org/p/liberty-nova-priorities | 00:17 |
| jogo | and go | 00:17 |
| *** hdd has quit IRC | 00:18 | |
| jogo | mikal dansmith: email posted | 00:20 |
| *** tsekiyam_ has joined #openstack-nova | 00:20 | |
| * dansmith closes his email | 00:20 | |
| jogo | hehe | 00:20 |
| dansmith | just kidding I never close it | 00:20 |
| dansmith | BUT | 00:20 |
| dansmith | I am going to go get food and not care about priorities until tomorrow | 00:20 |
| jogo | there is plenty of time to care about priorities later so good idea | 00:20 |
| jogo | mikal: it may be fun to look back at http://specs.openstack.org/openstack/nova-specs/priorities/kilo-priorities.html and see how well we did | 00:21 |
| mikal | Yeah, my inbox is full of email I don't really want to read right now | 00:21 |
| mikal | So I sympathize | 00:21 |
| *** mtanino has quit IRC | 00:22 | |
| jogo | cells2, sounds like its moving along a bit slowly | 00:22 |
| *** annegentle has quit IRC | 00:22 | |
| jogo | and I think no downtime db upgrades barely missed Kilo | 00:22 |
| *** tsekiyama has quit IRC | 00:22 | |
| *** Sukhdev has quit IRC | 00:23 | |
| dansmith | jogo: it did, and I was annoyed | 00:23 |
| dansmith | I tried to drum up reviews the last few days, but it just wasn't happening | 00:23 |
| jogo | objects made good progress right? | 00:23 |
| jogo | same for scheduler | 00:23 |
| dansmith | yes, both | 00:24 |
| dansmith | cells made good progress | 00:24 |
| dansmith | slow is expected given the gravity | 00:24 |
| jogo | dansmith: we should merge no db upgrades before the summit IMHO | 00:24 |
| jogo | v2.1 moved along really nicely too | 00:24 |
| dansmith | jogo: sure | 00:24 |
| jogo | same for functional testing | 00:24 |
| jogo | and CI | 00:24 |
| dansmith | jogo: or two weeks ago, either one | 00:24 |
| *** tsekiyam_ has quit IRC | 00:24 | |
| dansmith | I'm really leaving now | 00:24 |
| jogo | dansmith: o/ | 00:24 |
| jogo | me too o/ | 00:25 |
| *** jay-lau-513 has quit IRC | 00:25 | |
| *** zhangtralon has joined #openstack-nova | 00:30 | |
| *** sbfox has quit IRC | 00:34 | |
| *** yingjun has joined #openstack-nova | 00:35 | |
| *** marun has quit IRC | 00:43 | |
| *** jecarey has joined #openstack-nova | 00:55 | |
| *** Longgeek has joined #openstack-nova | 00:56 | |
| *** park_heijlong has quit IRC | 00:56 | |
| *** annegentle has joined #openstack-nova | 00:57 | |
| dims | jogo: mikal: this is missing a +W? https://review.openstack.org/#/c/170190/ | 00:57 |
| *** baoli has quit IRC | 00:57 | |
| *** vilobhmm1 has quit IRC | 00:59 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Merge evacuate functional tests between v2 and v2.1 https://review.openstack.org/171902 | 00:59 |
| *** tjones1 has joined #openstack-nova | 01:00 | |
| *** vilobhmm1 has joined #openstack-nova | 01:02 | |
| *** vilobhmm1 has quit IRC | 01:03 | |
| *** vilobhmm1 has joined #openstack-nova | 01:03 | |
| *** rfolco has quit IRC | 01:05 | |
| *** vilobhmm1 has quit IRC | 01:10 | |
| *** vilobhmm1 has joined #openstack-nova | 01:10 | |
| *** hdd has joined #openstack-nova | 01:12 | |
| *** yingjun has quit IRC | 01:12 | |
| *** vilobhmm1 has quit IRC | 01:15 | |
| *** Marga_ has quit IRC | 01:19 | |
| *** Marga_ has joined #openstack-nova | 01:20 | |
| *** mtanino has joined #openstack-nova | 01:21 | |
| *** annegentle has quit IRC | 01:22 | |
| *** hdd has quit IRC | 01:24 | |
| mikal | dims: looking | 01:26 |
| mikal | dims: ahhh sorry, that was me having a shitty day yesterday and forgotting. Fixed. | 01:27 |
| *** mriedem1 has joined #openstack-nova | 01:28 | |
| *** mriedem has quit IRC | 01:28 | |
| *** yamahata has quit IRC | 01:28 | |
| *** setuid has quit IRC | 01:28 | |
| *** asettle is now known as asettle-gymk | 01:29 | |
| *** asettle-gymk is now known as asettle-gym | 01:29 | |
| dims | mikal: thanks for the moving email! still in shock! | 01:29 |
| mikal | dims: yeah, its pretty shit | 01:29 |
| mikal | dims: such is life though I suppose | 01:29 |
| dims | :( | 01:30 |
| *** setuid has joined #openstack-nova | 01:32 | |
| *** yingjun has joined #openstack-nova | 01:34 | |
| *** dboik_ has quit IRC | 01:34 | |
| *** jay-lau-513 has joined #openstack-nova | 01:40 | |
| *** baoli has joined #openstack-nova | 01:45 | |
| *** otter768 has joined #openstack-nova | 01:49 | |
| *** annegentle has joined #openstack-nova | 01:52 | |
| *** otter768 has quit IRC | 01:53 | |
| *** zhangtralon has quit IRC | 01:53 | |
| *** erkules_ has joined #openstack-nova | 01:56 | |
| *** dims has quit IRC | 01:56 | |
| *** erkules has quit IRC | 01:58 | |
| *** baoli has quit IRC | 01:59 | |
| *** tjones1 has quit IRC | 02:01 | |
| *** zzzeek_ has quit IRC | 02:02 | |
| *** harlowja is now known as harlowja_away | 02:03 | |
| *** browne has quit IRC | 02:07 | |
| *** park_heijlong has joined #openstack-nova | 02:10 | |
| *** baoli has joined #openstack-nova | 02:11 | |
| *** zhangtralon has joined #openstack-nova | 02:16 | |
| *** iamjarvo has joined #openstack-nova | 02:21 | |
| *** annegentle has quit IRC | 02:24 | |
| *** Marga_ has quit IRC | 02:26 | |
| *** isd has joined #openstack-nova | 02:26 | |
| *** Marga_ has joined #openstack-nova | 02:26 | |
| *** iamjarvo has quit IRC | 02:29 | |
| *** unicell has quit IRC | 02:30 | |
| *** Marga_ has quit IRC | 02:31 | |
| *** jogo has quit IRC | 02:34 | |
| openstackgerrit | melanie witt proposed openstack/nova: Honor uuid parameter passed to nova-network create https://review.openstack.org/171911 | 02:34 |
| *** yingjun has quit IRC | 02:37 | |
| *** browne has joined #openstack-nova | 02:38 | |
| *** yamahata has joined #openstack-nova | 02:40 | |
| *** annashen has joined #openstack-nova | 02:41 | |
| *** annashen has quit IRC | 02:42 | |
| *** annashen has joined #openstack-nova | 02:42 | |
| openstackgerrit | melanie witt proposed openstack/nova: Honor uuid parameter passed to nova-network create https://review.openstack.org/171911 | 02:43 |
| *** Longgeek has quit IRC | 02:43 | |
| openstackgerrit | melanie witt proposed openstack/nova: Refactor nova-net cidr validation in prep for bug fix https://review.openstack.org/171912 | 02:43 |
| *** Longgeek has joined #openstack-nova | 02:44 | |
| *** tjones1 has joined #openstack-nova | 02:46 | |
| *** mriedem1 is now known as mriedem_away | 02:48 | |
| *** zhangtralon has quit IRC | 02:50 | |
| *** VW_ has joined #openstack-nova | 02:50 | |
| *** zhangtralon has joined #openstack-nova | 02:53 | |
| *** asettle-gym is now known as asettle | 02:59 | |
| *** annashen has quit IRC | 03:07 | |
| *** alex_xu has quit IRC | 03:08 | |
| *** baoli has quit IRC | 03:09 | |
| *** baoli has joined #openstack-nova | 03:13 | |
| *** xyang1 has quit IRC | 03:13 | |
| *** annashen has joined #openstack-nova | 03:14 | |
| *** yingjun has joined #openstack-nova | 03:16 | |
| *** VW_ has quit IRC | 03:17 | |
| *** xuhanp has joined #openstack-nova | 03:18 | |
| *** xuhanp_ has joined #openstack-nova | 03:18 | |
| *** rushiagr_away is now known as rushiagr | 03:22 | |
| openstackgerrit | zhangtralon proposed openstack/nova-specs: Make fitler to support force_hosts/force_nodes https://review.openstack.org/171917 | 03:22 |
| *** VW_ has joined #openstack-nova | 03:23 | |
| *** dsanders has quit IRC | 03:23 | |
| *** VW_ has quit IRC | 03:24 | |
| *** Longgeek_ has joined #openstack-nova | 03:29 | |
| *** Longgeek has quit IRC | 03:30 | |
| *** sorrison has quit IRC | 03:30 | |
| *** sorrison has joined #openstack-nova | 03:31 | |
| *** erlon has quit IRC | 03:31 | |
| openstackgerrit | zhangtralon proposed openstack/nova-specs: Make fitler to support force_hosts/force_nodes https://review.openstack.org/171917 | 03:31 |
| *** tojuvone has joined #openstack-nova | 03:31 | |
| *** annashen has quit IRC | 03:32 | |
| *** ozialien has joined #openstack-nova | 03:35 | |
| openstackgerrit | zhangtralon proposed openstack/nova-specs: Make fitler to support force_hosts/force_nodes https://review.openstack.org/171917 | 03:35 |
| *** rushiagr is now known as rushiagr_away | 03:36 | |
| *** achanda has quit IRC | 03:38 | |
| *** dave-mccowan has quit IRC | 03:40 | |
| *** achanda has joined #openstack-nova | 03:43 | |
| *** dims has joined #openstack-nova | 03:45 | |
| *** achanda has quit IRC | 03:45 | |
| *** joefides has joined #openstack-nova | 03:46 | |
| *** Maike_ has quit IRC | 03:48 | |
| *** annashen has joined #openstack-nova | 03:48 | |
| *** Maike_ has joined #openstack-nova | 03:48 | |
| *** joefides_ has joined #openstack-nova | 03:48 | |
| *** otter768 has joined #openstack-nova | 03:49 | |
| *** jogo has joined #openstack-nova | 03:50 | |
| *** dims has quit IRC | 03:50 | |
| *** tjones1 has quit IRC | 03:50 | |
| *** annashen has quit IRC | 03:51 | |
| *** joefides has quit IRC | 03:51 | |
| *** tjones1 has joined #openstack-nova | 03:52 | |
| *** artom has quit IRC | 03:53 | |
| *** iamjarvo has joined #openstack-nova | 03:54 | |
| *** otter768 has quit IRC | 03:54 | |
| *** jogo has quit IRC | 03:55 | |
| *** yamahata has quit IRC | 03:55 | |
| *** yamahata has joined #openstack-nova | 03:55 | |
| *** foexle_ has joined #openstack-nova | 03:55 | |
| *** deepthi has joined #openstack-nova | 03:58 | |
| *** foexle has quit IRC | 03:58 | |
| *** baoli has quit IRC | 03:58 | |
| *** coolsvap|afk is now known as coolsvap | 04:00 | |
| *** achanda has joined #openstack-nova | 04:01 | |
| *** yingjun has quit IRC | 04:02 | |
| *** rushiagr_away is now known as rushiagr | 04:02 | |
| openstackgerrit | zhangtralon proposed openstack/nova-specs: Make fitler to support force_hosts/force_nodes https://review.openstack.org/171917 | 04:02 |
| openstackgerrit | zhangtralon proposed openstack/nova: add support for muti_statuses_query from migration https://review.openstack.org/169631 | 04:03 |
| *** achanda has quit IRC | 04:03 | |
| *** jogo has joined #openstack-nova | 04:05 | |
| *** rodrigods has quit IRC | 04:08 | |
| *** browne has quit IRC | 04:10 | |
| *** rodrigods has joined #openstack-nova | 04:12 | |
| openstackgerrit | zhangtralon proposed openstack/nova: resize-revert can break anti-affinity policy https://review.openstack.org/160110 | 04:12 |
| *** rushiagr is now known as rushiagr_away | 04:12 | |
| *** unicell has joined #openstack-nova | 04:13 | |
| *** rajesht has joined #openstack-nova | 04:21 | |
| *** achanda has joined #openstack-nova | 04:23 | |
| *** dsanders has joined #openstack-nova | 04:27 | |
| *** vilobhmm1 has joined #openstack-nova | 04:29 | |
| *** iamjarvo has quit IRC | 04:31 | |
| *** raginbajin has quit IRC | 04:32 | |
| *** isd has left #openstack-nova | 04:34 | |
| *** raginbajin has joined #openstack-nova | 04:35 | |
| *** VW_ has joined #openstack-nova | 04:36 | |
| *** achanda has quit IRC | 04:36 | |
| *** VW_ has quit IRC | 04:39 | |
| *** vivekd has joined #openstack-nova | 04:39 | |
| *** unicell has quit IRC | 04:42 | |
| *** veena has joined #openstack-nova | 04:43 | |
| *** browne has joined #openstack-nova | 04:46 | |
| *** melwitt has quit IRC | 04:47 | |
| *** rushiagr_away is now known as rushiagr | 04:47 | |
| *** achanda has joined #openstack-nova | 04:48 | |
| *** yingjun has joined #openstack-nova | 04:48 | |
| *** iamjarvo has joined #openstack-nova | 04:48 | |
| *** iamjarvo has quit IRC | 04:48 | |
| *** iamjarvo has joined #openstack-nova | 04:49 | |
| *** zhangtralon has quit IRC | 04:51 | |
| *** dsanders has quit IRC | 04:54 | |
| *** tjones1 has quit IRC | 04:55 | |
| *** xuhanp_ has quit IRC | 04:57 | |
| *** xuhanp has quit IRC | 04:58 | |
| *** rushiagr is now known as rushiagr_away | 04:58 | |
| *** iamjarvo has quit IRC | 05:00 | |
| *** vivekd_ has joined #openstack-nova | 05:04 | |
| *** yingjun has quit IRC | 05:04 | |
| *** vivekd has quit IRC | 05:04 | |
| *** xuhanp has joined #openstack-nova | 05:04 | |
| *** xuhanp_ has joined #openstack-nova | 05:04 | |
| *** vivekd_ is now known as vivekd | 05:04 | |
| *** ozialien has quit IRC | 05:06 | |
| *** dsanders has joined #openstack-nova | 05:13 | |
| *** dsanders has quit IRC | 05:21 | |
| *** kylek3h has quit IRC | 05:23 | |
| *** kylek3h has joined #openstack-nova | 05:23 | |
| *** _gryf|afk is now known as _gryf | 05:27 | |
| *** annashen has joined #openstack-nova | 05:34 | |
| *** ajayaa has joined #openstack-nova | 05:38 | |
| *** yingjun has joined #openstack-nova | 05:39 | |
| *** yingjun has quit IRC | 05:40 | |
| *** bkopilov has quit IRC | 05:41 | |
| *** irenab has joined #openstack-nova | 05:42 | |
| *** bkopilov has joined #openstack-nova | 05:43 | |
| *** arnaud___ has joined #openstack-nova | 05:44 | |
| openstackgerrit | Lingxian Kong proposed openstack/nova: Correct az property of instance when specifying force host/node https://review.openstack.org/163842 | 05:49 |
| *** ankit_ag has joined #openstack-nova | 05:49 | |
| *** otter768 has joined #openstack-nova | 05:50 | |
| *** wasmum has quit IRC | 05:52 | |
| *** otter768 has quit IRC | 05:55 | |
| *** park_heijlong has quit IRC | 05:57 | |
| *** jcoufal has joined #openstack-nova | 05:59 | |
| *** sahid has joined #openstack-nova | 06:00 | |
| *** browne has quit IRC | 06:00 | |
| *** yingjun has joined #openstack-nova | 06:04 | |
| *** ildikov has quit IRC | 06:04 | |
| *** lpetrut has joined #openstack-nova | 06:06 | |
| *** mtanino has quit IRC | 06:07 | |
| *** dulek has joined #openstack-nova | 06:07 | |
| *** oro has joined #openstack-nova | 06:08 | |
| *** park_heijlong has joined #openstack-nova | 06:11 | |
| *** arnaud___ has quit IRC | 06:15 | |
| *** yamahata has quit IRC | 06:16 | |
| *** annashen has quit IRC | 06:16 | |
| *** annashen has joined #openstack-nova | 06:18 | |
| *** yamahata has joined #openstack-nova | 06:20 | |
| *** sbfox has joined #openstack-nova | 06:26 | |
| *** zhangtralon has joined #openstack-nova | 06:29 | |
| *** rajesht has quit IRC | 06:31 | |
| *** sdake_ has joined #openstack-nova | 06:32 | |
| *** veena has quit IRC | 06:32 | |
| *** sbfox has quit IRC | 06:35 | |
| *** rajesht has joined #openstack-nova | 06:35 | |
| *** annashen has quit IRC | 06:36 | |
| *** dsanders has joined #openstack-nova | 06:38 | |
| *** park_heijlong has quit IRC | 06:38 | |
| *** romainh has joined #openstack-nova | 06:40 | |
| *** alex_xu has joined #openstack-nova | 06:40 | |
| *** rushiagr_away is now known as rushiagr | 06:43 | |
| *** alex_xu has quit IRC | 06:47 | |
| *** yamahata has quit IRC | 06:48 | |
| *** alex_xu has joined #openstack-nova | 06:51 | |
| *** park_heijlong has joined #openstack-nova | 06:51 | |
| *** vilobhmm1 has quit IRC | 06:51 | |
| *** annashen has joined #openstack-nova | 06:52 | |
| openstackgerrit | zhangtralon proposed openstack/nova-specs: Make fitler to support force_hosts/force_nodes https://review.openstack.org/171917 | 06:53 |
| *** annashen has quit IRC | 06:54 | |
| *** vilobhmm1 has joined #openstack-nova | 06:57 | |
| *** wasmum has joined #openstack-nova | 06:58 | |
| *** vilobhmm1 has quit IRC | 06:58 | |
| *** annashen has joined #openstack-nova | 06:58 | |
| *** jichen has joined #openstack-nova | 07:00 | |
| *** atuvenie has joined #openstack-nova | 07:01 | |
| *** markus_z has joined #openstack-nova | 07:03 | |
| *** sdake_ has quit IRC | 07:05 | |
| *** dsanders has quit IRC | 07:06 | |
| *** afazekas has joined #openstack-nova | 07:13 | |
| *** armax has quit IRC | 07:13 | |
| *** dsanders has joined #openstack-nova | 07:14 | |
| *** annashen has quit IRC | 07:14 | |
| *** jistr has joined #openstack-nova | 07:15 | |
| *** doude has joined #openstack-nova | 07:15 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Imported Translations from Transifex https://review.openstack.org/171943 | 07:15 |
| *** salv-orlando has joined #openstack-nova | 07:17 | |
| *** bkopilov has quit IRC | 07:18 | |
| *** sdake has joined #openstack-nova | 07:18 | |
| *** yingjun has quit IRC | 07:20 | |
| *** ildikov has joined #openstack-nova | 07:22 | |
| *** zhangtralon has quit IRC | 07:24 | |
| *** matrohon has joined #openstack-nova | 07:24 | |
| *** annashen has joined #openstack-nova | 07:24 | |
| *** sdake_ has joined #openstack-nova | 07:25 | |
| *** sdake has quit IRC | 07:26 | |
| *** salv-orlando has quit IRC | 07:27 | |
| *** salv-orlando has joined #openstack-nova | 07:28 | |
| *** sdake_ has quit IRC | 07:30 | |
| *** salv-orl_ has joined #openstack-nova | 07:30 | |
| hanzhang | hi all, who can help review this patch:https://review.openstack.org/#/c/171459/ | 07:30 |
| *** salv-orlando has quit IRC | 07:31 | |
| *** achanda has quit IRC | 07:31 | |
| *** gszasz has joined #openstack-nova | 07:34 | |
| *** atuvenie has quit IRC | 07:35 | |
| *** raph_ has quit IRC | 07:36 | |
| *** annashen has quit IRC | 07:36 | |
| *** achanda has joined #openstack-nova | 07:37 | |
| *** vivekd has quit IRC | 07:38 | |
| *** dsanders has quit IRC | 07:40 | |
| *** yingjun has joined #openstack-nova | 07:42 | |
| openstackgerrit | zhangtralon proposed openstack/nova: Support force_hosts in the ChanceScheduler https://review.openstack.org/171951 | 07:42 |
| *** oro has quit IRC | 07:46 | |
| *** achanda has quit IRC | 07:48 | |
| *** raph_ has joined #openstack-nova | 07:49 | |
| *** oomichi has joined #openstack-nova | 07:50 | |
| *** zhangtralon has joined #openstack-nova | 07:50 | |
| *** salv-orl_ has quit IRC | 07:50 | |
| *** otter768 has joined #openstack-nova | 07:51 | |
| *** sdake has joined #openstack-nova | 07:52 | |
| *** otter768 has quit IRC | 07:56 | |
| *** park_heijlong has quit IRC | 08:00 | |
| *** jamielennox is now known as jamielennox|away | 08:01 | |
| *** jlanoux has joined #openstack-nova | 08:04 | |
| *** oomichi has quit IRC | 08:06 | |
| openstackgerrit | sahid proposed openstack/nova: libvirt: quit early when mempages requested found https://review.openstack.org/171502 | 08:08 |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Merge extended server Att tests between v2 and v2.1 https://review.openstack.org/171961 | 08:09 |
| bauzas | mikal: morning | 08:09 |
| bauzas | erm, evening even | 08:09 |
| bauzas | mikal: re: your RC1 email, happy with discussing about cells job status ? | 08:10 |
| bauzas | mikal: I tried to clarify the Launchpad bugs | 08:10 |
| *** derekh has joined #openstack-nova | 08:13 | |
| *** erkules_ is now known as erkules | 08:14 | |
| *** erkules has joined #openstack-nova | 08:14 | |
| *** park_heijlong has joined #openstack-nova | 08:16 | |
| *** atuvenie has joined #openstack-nova | 08:16 | |
| *** apuimedo has joined #openstack-nova | 08:17 | |
| *** jasondotstar has quit IRC | 08:20 | |
| *** lucasagomes has joined #openstack-nova | 08:24 | |
| *** andreykurilin_ has joined #openstack-nova | 08:27 | |
| *** pradipta has joined #openstack-nova | 08:28 | |
| *** dheeraj-gupta-4 has joined #openstack-nova | 08:29 | |
| *** dheeraj-gupta-4 has quit IRC | 08:29 | |
| *** ildikov has quit IRC | 08:30 | |
| *** jlanoux has quit IRC | 08:30 | |
| *** jlanoux has joined #openstack-nova | 08:34 | |
| *** zz_johnthetubagu is now known as johnthetubaguy | 08:34 | |
| *** oro has joined #openstack-nova | 08:36 | |
| *** doude has quit IRC | 08:36 | |
| *** doude has joined #openstack-nova | 08:36 | |
| *** heyongli has joined #openstack-nova | 08:38 | |
| openstackgerrit | Lingxian Kong proposed openstack/nova: Correct az property of instance when specifying force host/node https://review.openstack.org/163842 | 08:39 |
| *** rwsu has quit IRC | 08:41 | |
| *** jyuso has joined #openstack-nova | 08:43 | |
| *** ildikov has joined #openstack-nova | 08:45 | |
| johnthetubaguy | bauzas: just looking at the RC1 bug list, and the cells patch | 08:45 |
| bauzas | johnthetubaguy: I made a status on the ML | 08:46 |
| bauzas | johnthetubaguy: http://lists.openstack.org/pipermail/openstack-dev/2015-April/061039.html | 08:46 |
| johnthetubaguy | bauzas: ah, OK, I should go read that... | 08:46 |
| *** ndipanov has joined #openstack-nova | 08:46 | |
| bauzas | johnthetubaguy: I just tried to put critical and set for RC1, but that's up to be discussed, for sure :) | 08:46 |
| johnthetubaguy | bauzas: well, I think its more high, due to the wideness of impact, but we can try and get it into RC1, while there are other blockers | 08:49 |
| bauzas | johnthetubaguy: np, changing the prio | 08:49 |
| johnthetubaguy | bauzas: I am thinking this needs more unit tests though: https://review.openstack.org/#/c/160506/ | 08:49 |
| bauzas | johnthetubaguy: oh, you beated me :) | 08:49 |
| bauzas | johnthetubaguy: oh,you missed the cell test | 08:50 |
| johnthetubaguy | bauzas: ah, I wondered if I did, where is it? | 08:50 |
| bauzas | johnthetubaguy: https://review.openstack.org/#/c/160506/20/nova/tests/functional/test_api_samples.py,cm | 08:50 |
| bauzas | bug agreed, that's a func test | 08:51 |
| bauzas | but* | 08:51 |
| *** doude has quit IRC | 08:51 | |
| *** lpetrut has quit IRC | 08:51 | |
| johnthetubaguy | bauzas: hmm, OK… I wondered about a subclass for the API unit tests that stubs out the cells rpcapi, or something like that? | 08:51 |
| bauzas | johnthetubaguy: sounds doable to proivide a TestCase of https://review.openstack.org/#/c/160506/20/nova/tests/unit/api/openstack/compute/contrib/test_hypervisors.py,cm | 08:51 |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Merge extended server status tests between v2 and v2.1 https://review.openstack.org/171970 | 08:51 |
| bauzas | related to cells | 08:51 |
| johnthetubaguy | bauzas: right, thats the one | 08:52 |
| bauzas | johnthetubaguy: sure, can do that, lemme try | 08:52 |
| *** doude has joined #openstack-nova | 08:52 | |
| johnthetubaguy | bauzas: thanks, if its not maintainable, then thats probably cool, just wondering how explicit we could go | 08:52 |
| bauzas | johnthetubaguy: sure thing | 08:52 |
| johnthetubaguy | bauzas: I could see someone just reverting the host_api change, saying, not need for this, call object directly again | 08:53 |
| johnthetubaguy | a unit test to stop that would be cool | 08:53 |
| *** Maike_ has quit IRC | 08:54 | |
| bauzas | johnthetubaguy: yeah make sense | 08:54 |
| bauzas | johnthetubaguy: I was actuall pro bypassing the hostapi and provide an cell-aware object | 08:54 |
| bauzas | johnthetubaguy: but we sucked because of the serialization of those Proxy objects | 08:54 |
| johnthetubaguy | bauzas: ah, gotcha, thanks | 08:55 |
| *** zhangtralon has quit IRC | 08:55 | |
| *** eglynn has joined #openstack-nova | 09:09 | |
| *** park_heijlong has quit IRC | 09:10 | |
| *** boris-42 has joined #openstack-nova | 09:16 | |
| *** pcaruana has joined #openstack-nova | 09:17 | |
| *** coolsvap is now known as coolsvap|afk | 09:18 | |
| *** bkopilov has joined #openstack-nova | 09:19 | |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: scheduler: re-calculate NUMA on consume_from_instance https://review.openstack.org/169245 | 09:19 |
| *** claudiub has joined #openstack-nova | 09:23 | |
| *** coolsvap|afk is now known as coolsvap | 09:24 | |
| johnthetubaguy | ndipanov: do we want to block the release on that bug, BTW, seems like something that we could backport easily, or am I missing something? | 09:24 |
| *** lpetrut has joined #openstack-nova | 09:24 | |
| ndipanov | johnthetubaguy, on that one... maybe not | 09:25 |
| ndipanov | but there is another one | 09:25 |
| ndipanov | that I'd like fixed for RC1 | 09:25 |
| johnthetubaguy | ndipanov: ah, interesting, which one is that? | 09:25 |
| ndipanov | was just about to reply to mikal s email | 09:25 |
| * ndipanov digs | 09:25 | |
| ndipanov | johnthetubaguy, also - https://review.openstack.org/#/c/169752/ would be really nice to have in kilo | 09:25 |
| ndipanov | but that's not the one | 09:26 |
| ndipanov | johnthetubaguy, https://review.openstack.org/#/c/131321 | 09:27 |
| ndipanov | not sure how that slipped back to high prio from critical | 09:27 |
| ndipanov | it's pretty bad imho | 09:27 |
| ndipanov | and the patch is close enough I think | 09:27 |
| ndipanov | but need to look at it in detail | 09:27 |
| ndipanov | which in case of that code will take 1h + | 09:27 |
| ndipanov | johnthetubaguy, ah I see how :) | 09:28 |
| johnthetubaguy | ndipanov: hmm, so that one got dropped as it didn't seem like we would ever merge anything in time | 09:28 |
| johnthetubaguy | ndipanov: agreed its important though | 09:28 |
| ndipanov | let me review it now | 09:28 |
| ndipanov | would it be more desirable to cur rc1 now and do an RC2 if this makes it? | 09:30 |
| ndipanov | or do we hate rc2 as usual :) | 09:30 |
| ndipanov | thinking about how backportable this is... | 09:31 |
| ndipanov | it's probably not too bad | 09:31 |
| ndipanov | the websocket proxy config one is the one I'd say we should merge | 09:31 |
| *** yingjun has quit IRC | 09:31 | |
| ndipanov | for PCI - you're right, we can ship without it | 09:32 |
| ndipanov | johnthetubaguy, ^ | 09:32 |
| johnthetubaguy | ndipanov: RC2 is a faf, but I like the idea of only cutting RC2 if that patch makes it | 09:32 |
| johnthetubaguy | ndipanov: can you add that idea on that bug report, that would be great | 09:33 |
| ndipanov | sure | 09:33 |
| ndipanov | do consider the | 09:33 |
| ndipanov | https://review.openstack.org/#/c/169752/ though | 09:33 |
| ndipanov | it's got +2s from alaski who was one of the sponsors of the original fix | 09:34 |
| johnthetubaguy | ndipanov: yeah, I am just reading through it, 52 looks good, just looking at the -1 on the following patch | 09:35 |
| *** yingjun has joined #openstack-nova | 09:35 | |
| ndipanov | oh I missed that | 09:35 |
| johnthetubaguy | ndipanov: i think you actually have the test she wan't already | 09:36 |
| johnthetubaguy | wants^ | 09:36 |
| *** deepthi has quit IRC | 09:37 | |
| *** yingjun has quit IRC | 09:37 | |
| ndipanov | yep - 2 tests above | 09:37 |
| *** ildikov has quit IRC | 09:39 | |
| johnthetubaguy | ndipanov: so this one: https://review.openstack.org/#/c/169753 bit worried about upgrade | 09:39 |
| johnthetubaguy | not sure if I read the code correctly though | 09:39 |
| ndipanov | reading | 09:39 |
| *** haigang has joined #openstack-nova | 09:39 | |
| ndipanov | hmm that is a good point | 09:40 |
| johnthetubaguy | I mean we could leave that nasty code that needs the config, and delete it when liberty opens, but yeah… | 09:41 |
| johnthetubaguy | not sure that really helps people either, given your valid comment around the config | 09:41 |
| johnthetubaguy | ndipanov: do we know the version of the client calling in that method, I guess not? | 09:41 |
| ndipanov | we do | 09:42 |
| ndipanov | well | 09:42 |
| ndipanov | we don't actually | 09:42 |
| ndipanov | as the webcosket client will be new | 09:42 |
| ndipanov | we would have to have the API pass in a flag | 09:43 |
| ndipanov | that says | 09:43 |
| *** ZZelle has quit IRC | 09:43 | |
| ndipanov | when you validate this, dear proxy | 09:43 |
| ndipanov | don't check for access url there | 09:43 |
| ndipanov | but that defeats the purpose | 09:43 |
| johnthetubaguy | right | 09:43 |
| *** ZZelle has joined #openstack-nova | 09:43 | |
| ndipanov | hmmm | 09:43 |
| openstackgerrit | wanghao proposed openstack/nova: Allow bdm boot_index=1 when creating vm from image https://review.openstack.org/171984 | 09:43 |
| johnthetubaguy | so part of me wants to say, its a security issues, life sucks, move on | 09:43 |
| ndipanov | and yar - assuming juno-kilo upgrade | 09:44 |
| ndipanov | anyone would have to have juno running with config options on proxy hosts | 09:44 |
| johnthetubaguy | so I guess its do we fail closed, or fail open | 09:44 |
| ndipanov | ideally - kilo would need to have both ways to validate | 09:45 |
| ndipanov | and then L can move away from config | 09:45 |
| johnthetubaguy | yeah | 09:45 |
| johnthetubaguy | thats the safe fix | 09:45 |
| ndipanov | well that will take me some time | 09:46 |
| johnthetubaguy | maybe mark the config as deprecated | 09:46 |
| ndipanov | well it's deprecated only on proxies - computes still need it | 09:46 |
| johnthetubaguy | ndipanov: oh crud, yes | 09:46 |
| *** sdake_ has joined #openstack-nova | 09:46 | |
| ndipanov | we could relnote that you can't upgrade websockify to kilo until all of your computes are new | 09:47 |
| johnthetubaguy | ndipanov: yuck… so in good news, because I +Wed the API change, its possible we have the new API in liberty, so can then just drop the config there? | 09:47 |
| ndipanov | that's really waht would fix it - right? | 09:47 |
| johnthetubaguy | ndipanov: yes, that might be OK, with the pinning in place | 09:47 |
| ndipanov | I'd wait for dansmith on that one though | 09:48 |
| ndipanov | in reality as we've learned | 09:48 |
| johnthetubaguy | ndipanov: if you screw that up, you just get that exception, actually maybe thats OK ish | 09:48 |
| ndipanov | yeah | 09:48 |
| ndipanov | I mean | 09:48 |
| johnthetubaguy | ndipanov: maybe just update with UpgradeImpact about that for now, and as you say, chat with Dan when he comes online | 09:48 |
| ndipanov | ok | 09:49 |
| *** deepthi has joined #openstack-nova | 09:49 | |
| ndipanov | updating now | 09:49 |
| ndipanov | good catch | 09:49 |
| ndipanov | thanks! | 09:49 |
| *** sdake_ has quit IRC | 09:49 | |
| johnthetubaguy | ndipanov: np | 09:49 |
| johnthetubaguy | ndipanov: just looking at the bug its linked against, stupid process thing, but we probably need another bug to track that patch as the other one is already "released" | 09:50 |
| *** sdake has quit IRC | 09:50 | |
| ndipanov | I can do that | 09:51 |
| ndipanov | ah so you tagged it | 09:51 |
| * ndipanov pulls | 09:51 | |
| ndipanov | hmmm | 09:51 |
| ndipanov | ah right | 09:52 |
| johnthetubaguy | that other bug is in a messy state | 09:52 |
| *** otter768 has joined #openstack-nova | 09:52 | |
| ndipanov | it was a security fix | 09:52 |
| johnthetubaguy | yeah | 09:52 |
| ndipanov | ok adding the bug and upgrade impact | 09:52 |
| johnthetubaguy | embargoed and all that | 09:52 |
| johnthetubaguy | ndipanov: sweet, thanks | 09:53 |
| *** ildikov has joined #openstack-nova | 09:54 | |
| *** f13o has quit IRC | 09:54 | |
| johnthetubaguy | bauzas: so looking at this one: https://bugs.launchpad.net/nova/+bug/1441931 I am ok it getting the potential tag, but feels bad blocking rc1 on this? | 09:54 |
| openstack | Launchpad bug 1441931 in OpenStack Compute (nova) "nova-network create ignores uuid request parameter" [Critical,In progress] - Assigned to melanie witt (melwitt) | 09:55 |
| bauzas | johnthetubaguy: humpf, it depends on if we want to merge it for RC1 :) | 09:55 |
| bauzas | johnthetubaguy: I mean, if we want the cells job green for RC1 | 09:56 |
| johnthetubaguy | bauzas: I don't think we should block the release on that, its too late | 09:56 |
| bauzas | johnthetubaguy: planning to do RC2 ? | 09:56 |
| ndipanov | johnthetubaguy, rc potential or tag? | 09:56 |
| johnthetubaguy | bauzas: not really, but it might happen if something really bad comes up | 09:56 |
| ndipanov | tagged | 09:56 |
| ndipanov | https://bugs.launchpad.net/nova/+bug/1442048 johnthetubaguy | 09:57 |
| openstack | Launchpad bug 1442048 in OpenStack Compute (nova) "Avoid websocket proxies needing to have matching have config '*_baseurl' configs with compute nodes" [Undecided,Confirmed] | 09:57 |
| *** otter768 has quit IRC | 09:57 | |
| bauzas | johnthetubaguy: okay, when do you plan to tag rc1 ? | 09:57 |
| bauzas | johnthetubaguy: I mean, today EOB or before ? | 09:57 |
| johnthetubaguy | bauzas: should be today ish, if possible | 09:57 |
| bauzas | johnthetubaguy: yeah, ok, let's wait alaski if possible for this | 09:58 |
| johnthetubaguy | thats fine | 09:58 |
| bauzas | johnthetubaguy: because we agreed on trying to merge if possible by RC1 | 09:58 |
| johnthetubaguy | bauzas: I agree, but that meant it all being merged yesterday, and its not | 09:58 |
| bauzas | johnthetubaguy: agreed | 09:58 |
| bauzas | johnthetubaguy: we're late | 09:58 |
| bauzas | I haven't looked at Zuul yet for today | 09:59 |
| bauzas | lemme check the current delay | 09:59 |
| bauzas | umpf | 09:59 |
| bauzas | the gate pipeline is quite good, but the check queue is growing | 10:00 |
| *** vivekd has joined #openstack-nova | 10:03 | |
| johnthetubaguy | ndipanov: did you see this one: https://review.openstack.org/#/c/171304/ wondering if thats a big deal | 10:04 |
| ndipanov | johnthetubaguy, looking | 10:04 |
| ndipanov | hmmm | 10:04 |
| *** jcoufal has quit IRC | 10:04 | |
| ndipanov | I thought we actually tested this and it works | 10:04 |
| ndipanov | anyway - thinking about the config thing some more - actually everything will be fine - no upgrade impact | 10:06 |
| *** coolsvap is now known as coolsvap|afk | 10:06 | |
| ndipanov | the thing that sends the initial authorize req | 10:06 |
| ndipanov | is the API | 10:06 |
| ndipanov | and API is always new | 10:06 |
| ndipanov | so as long as you upgrade the API to kilo | 10:06 |
| ndipanov | you're good no matter what the compute nodes say | 10:06 |
| ndipanov | johnthetubaguy, ^^ | 10:07 |
| johnthetubaguy | ndipanov: arse, we say API last though: http://docs.openstack.org/developer/nova/devref/upgrade.html | 10:07 |
| johnthetubaguy | ndipanov: but you are right, we could just say API needs to be kilo in this case | 10:07 |
| ndipanov | hmm that document does not seem to be super precise - you have to roll conductors with the DB + you can't really keep API and sched back without flipping the db switch | 10:10 |
| ndipanov | but it's confusing enough to need a note for sure | 10:10 |
| johnthetubaguy | ndipanov: its assuming we merge the online migration code | 10:10 |
| johnthetubaguy | ndipanov: that way its no longer lock step | 10:10 |
| *** deepthi has quit IRC | 10:11 | |
| ndipanov | yeah wasn't following that closely - but I assume it's not in? so juno->kilo won't benefit from it? | 10:12 |
| johnthetubaguy | ndipanov: correct, its forward looking | 10:13 |
| *** jay-lau-513 has quit IRC | 10:14 | |
| *** atuvenie has quit IRC | 10:21 | |
| *** IanGovett has joined #openstack-nova | 10:22 | |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: websocketproxy: Make protocol validation use connection_info https://review.openstack.org/169753 | 10:23 |
| ndipanov | johnthetubaguy, ^^ | 10:23 |
| *** dims_ has joined #openstack-nova | 10:23 | |
| johnthetubaguy | ndipanov: I think there is an UpgradeImpact tag thats worth add? | 10:23 |
| ndipanov | arhg I said Upgrade-Impact | 10:24 |
| johnthetubaguy | I should check that... | 10:24 |
| ndipanov | yar | 10:25 |
| ndipanov | http://lists.openstack.org/pipermail/openstack-dev/2013-October/016665.html | 10:25 |
| johnthetubaguy | https://wiki.openstack.org/wiki/GitCommitMessages | 10:26 |
| johnthetubaguy | yeah | 10:26 |
| ndipanov | respinning | 10:26 |
| *** atuvenie has joined #openstack-nova | 10:26 | |
| rajesht | jogo: you around ? | 10:26 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: websocketproxy: Make protocol validation use connection_info https://review.openstack.org/169753 | 10:26 |
| *** deepthi has joined #openstack-nova | 10:28 | |
| *** jcoufal has joined #openstack-nova | 10:29 | |
| *** jcoufal has quit IRC | 10:36 | |
| *** annashen has joined #openstack-nova | 10:37 | |
| *** vivekd has quit IRC | 10:38 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds VolumeOps unit tests https://review.openstack.org/139797 | 10:40 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds MigrationOps unit tests https://review.openstack.org/139796 | 10:40 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: removes test_hypervapi tests https://review.openstack.org/139798 | 10:40 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds HyperVDriver unit tests https://review.openstack.org/148980 | 10:40 |
| *** jyuso has quit IRC | 10:41 | |
| *** annashen has quit IRC | 10:41 | |
| *** jamielennox|away is now known as jamielennox | 10:42 | |
| *** aix has joined #openstack-nova | 10:51 | |
| openstackgerrit | Rajesh Tailor proposed openstack/nova: Remove cell api overrides for force-delete https://review.openstack.org/135202 | 10:57 |
| *** ociuhandu has joined #openstack-nova | 10:57 | |
| *** hdd has joined #openstack-nova | 11:04 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Fix how service objects are looked up for Cells https://review.openstack.org/160506 | 11:07 |
| bauzas | johnthetubaguy: re:, I updated the patch with your comment ^ | 11:09 |
| *** asettle has quit IRC | 11:11 | |
| *** eglynn is now known as eglynn-brb | 11:12 | |
| johnthetubaguy | bauzas: thanks for that, looks good | 11:15 |
| bauzas | johnthetubaguy: np | 11:15 |
| bauzas | johnthetubaguy: thanks for quickly +2'ing it | 11:15 |
| bauzas | now I have a long mikal's email to reply... :) | 11:16 |
| *** johnthetubaguy is now known as zz_johnthetubagu | 11:16 | |
| *** subscope has quit IRC | 11:17 | |
| *** subscope has joined #openstack-nova | 11:20 | |
| *** andreykurilin_ has quit IRC | 11:20 | |
| *** doude has quit IRC | 11:23 | |
| *** haigang has quit IRC | 11:24 | |
| *** doude has joined #openstack-nova | 11:24 | |
| *** asettle has joined #openstack-nova | 11:25 | |
| *** lucasagomes is now known as lucas-hungry | 11:27 | |
| *** xuhanp has quit IRC | 11:30 | |
| *** xuhanp_ has quit IRC | 11:30 | |
| *** atuvenie has quit IRC | 11:36 | |
| *** tojuvone has quit IRC | 11:36 | |
| *** eliqiao has quit IRC | 11:40 | |
| openstackgerrit | wanghao proposed openstack/nova: fix 'Ignoring EndpointNotFound' error in init host https://review.openstack.org/147042 | 11:45 |
| *** lucas-hungry is now known as lucasagomes | 11:48 | |
| *** jamielennox is now known as jamielennox|away | 11:53 | |
| *** otter768 has joined #openstack-nova | 11:53 | |
| *** baoli has joined #openstack-nova | 11:55 | |
| *** baoli has quit IRC | 11:56 | |
| *** baoli has joined #openstack-nova | 11:57 | |
| *** pradipta has quit IRC | 11:57 | |
| *** otter768 has quit IRC | 11:58 | |
| hanzhang | hi all, who can help review this patch:https://review.openstack.org/#/c/171459/ | 12:01 |
| *** eglynn-brb is now known as eglynn | 12:04 | |
| *** sks has joined #openstack-nova | 12:05 | |
| *** RuiChen has quit IRC | 12:06 | |
| *** rodrigods has quit IRC | 12:06 | |
| *** rodrigods has joined #openstack-nova | 12:06 | |
| *** dprince has joined #openstack-nova | 12:07 | |
| *** RuiChen has joined #openstack-nova | 12:07 | |
| sks | I am using the devstack and after that when I tried nova boot from the image in the glance it fails. Can anybody please help? http://paste.openstack.org/show/201227/ | 12:07 |
| openstackgerrit | Abhishek Kekane proposed openstack/nova: Copy instance files from source to dest node https://review.openstack.org/150337 | 12:08 |
| openstackgerrit | Abhishek Kekane proposed openstack/nova: Add new HostAggregateGroupFilter scheduler filter https://review.openstack.org/150330 | 12:08 |
| openstackgerrit | Abhishek Kekane proposed openstack/nova: Move image creation from compute api to manager https://review.openstack.org/150315 | 12:08 |
| openstackgerrit | Abhishek Kekane proposed openstack/nova: Improve performance of Unshelve api https://review.openstack.org/150344 | 12:08 |
| *** dims_ has quit IRC | 12:10 | |
| *** jecarey has quit IRC | 12:12 | |
| *** dims_ has joined #openstack-nova | 12:13 | |
| *** liusheng has quit IRC | 12:15 | |
| openstackgerrit | Abhishek Kekane proposed openstack/nova: Add new HostAggregateGroupFilter scheduler filter https://review.openstack.org/150330 | 12:17 |
| openstackgerrit | Abhishek Kekane proposed openstack/nova: Move image creation from compute api to manager https://review.openstack.org/150315 | 12:17 |
| *** mtanino has joined #openstack-nova | 12:18 | |
| openstackgerrit | Abhishek Kekane proposed openstack/nova: Copy instance files from source to dest node https://review.openstack.org/150337 | 12:18 |
| *** aswadr has joined #openstack-nova | 12:20 | |
| *** aysyd has joined #openstack-nova | 12:22 | |
| *** manous has joined #openstack-nova | 12:22 | |
| manous | hi | 12:22 |
| *** tojuvone has joined #openstack-nova | 12:23 | |
| *** dims__ has joined #openstack-nova | 12:23 | |
| *** jyuso1 has joined #openstack-nova | 12:24 | |
| *** adalbas has joined #openstack-nova | 12:25 | |
| *** raildo|away is now known as raildo | 12:25 | |
| *** dims_ has quit IRC | 12:27 | |
| *** eharney has quit IRC | 12:27 | |
| *** asrangne has joined #openstack-nova | 12:27 | |
| *** deepthi has quit IRC | 12:28 | |
| *** joefides_ has quit IRC | 12:28 | |
| *** aswadr has quit IRC | 12:30 | |
| *** atuvenie has joined #openstack-nova | 12:31 | |
| *** hdd has quit IRC | 12:31 | |
| *** asrangne has quit IRC | 12:31 | |
| *** aswadr has joined #openstack-nova | 12:33 | |
| *** sudipto has joined #openstack-nova | 12:36 | |
| sudipto | dansmith, Hi, A quick question on fixed_ips table in nova. I am guessing that's used for nova-network and would be left empty for neutron based deployments? | 12:36 |
| mriedem_away | sudipto: yes | 12:37 |
| *** browne has joined #openstack-nova | 12:40 | |
| *** rfolco has joined #openstack-nova | 12:41 | |
| *** zz_johnthetubagu is now known as johnthetubaguy | 12:41 | |
| sudipto | mriedem_away, thanks! | 12:42 |
| *** mriedem_away is now known as mriedem | 12:42 | |
| *** ozialien has joined #openstack-nova | 12:43 | |
| ndipanov | dansmith, this one is waiting on you once you're caffeinated | 12:44 |
| ndipanov | https://review.openstack.org/#/c/169753/ | 12:44 |
| ndipanov | would be great to get it into kilo | 12:44 |
| *** hanzhang has quit IRC | 12:44 | |
| *** hanzhang has joined #openstack-nova | 12:44 | |
| johnthetubaguy | ndipanov: bauzas: I am looking at this and thinking its not going to merge in the next day, but it might merge by the beginning of next week? https://review.openstack.org/#/c/131321 | 12:44 |
| johnthetubaguy | ndipanov: bauzas: just if thats true, we should really start cutting RC1 today… but open to other ideas | 12:45 |
| ndipanov | johnthetubaguy, well - I'm being super thorough - mostly it;s nits on there | 12:45 |
| ndipanov | but yeah I assumed you were going to go ahead and do it and be lenient about cutting rc2 if this lands | 12:46 |
| johnthetubaguy | ndipanov: I am cool with it triggering an RC2, just feels like we need to cut soon | 12:46 |
| johnthetubaguy | cools | 12:46 |
| ndipanov | that's fine with me | 12:46 |
| johnthetubaguy | thanks, just making sure I am not missing something | 12:46 |
| johnthetubaguy | dansmith: sdague: ndipanov: if we are cool cutting RC1 I think we have to merge this patch: https://review.openstack.org/#/c/171078/ | 12:48 |
| ndipanov | well I'd like to get the config thing in first ... sdague dansmith :) | 12:48 |
| sdague | johnthetubaguy: you have to set the branch first, right? | 12:48 |
| johnthetubaguy | sdague: I think the branch gets cut after that patch, leaving that stuff to ttx | 12:49 |
| sdague | johnthetubaguy: no, the branch needs to happen first | 12:49 |
| johnthetubaguy | hmm, OK | 12:49 |
| sdague | I was pretty sure at least | 12:50 |
| sudipto | mriedem, I figured that a bug is invalid. Do i need to wait for the submitter to mark it or I can mark it myself? | 12:50 |
| sdague | ndipanov: +A on the above patch | 12:50 |
| *** jyuso1 has quit IRC | 12:52 | |
| *** ildikov has quit IRC | 12:53 | |
| johnthetubaguy | dansmith: I guess we need the placeholder DB migrations before we do RC1, did you see anyone put that up for review yet? | 12:53 |
| *** aix has quit IRC | 12:53 | |
| *** aswadr has quit IRC | 12:54 | |
| johnthetubaguy | dansmith: was there RPC and object version stuff we wanted to do this time around as well? | 12:54 |
| sdague | johnthetubaguy: oh right, and that | 12:54 |
| *** _gryf has quit IRC | 12:55 | |
| *** aswadr has joined #openstack-nova | 12:55 | |
| *** dave-mccowan has joined #openstack-nova | 12:56 | |
| ndipanov | johnthetubaguy, RPC stuff comes after - right? | 13:00 |
| *** toan-tran has joined #openstack-nova | 13:00 | |
| ndipanov | thanks sdague | 13:00 |
| johnthetubaguy | ndipanov: I remember there being stuff before and stuff after, but I could be wrong, we have it written down on the wiki somewhere I think | 13:01 |
| ndipanov | johnthetubaguy, we should do that for sure | 13:01 |
| mriedem | bauzas: question in https://review.openstack.org/#/c/160506/ | 13:01 |
| *** mtanino has quit IRC | 13:01 | |
| mriedem | bauzas: you made a comment that the cells job was failing on that change b/c we need to fix the tempest/network uuid stuff, which melwitt has patches up for to nova and devstack | 13:02 |
| mriedem | bauzas: so i'm wondering if your host api patch should depend on melwitt's nova-network patch, and her patch should depend on the devstack change | 13:02 |
| mriedem | ? | 13:02 |
| mriedem | then if the planets are all aligned, the cells job will pass on your change at the very end | 13:02 |
| *** toan-tran has joined #openstack-nova | 13:02 | |
| *** aix has joined #openstack-nova | 13:03 | |
| *** joefides has joined #openstack-nova | 13:04 | |
| *** pkoniszewski has joined #openstack-nova | 13:05 | |
| *** whenry has joined #openstack-nova | 13:06 | |
| *** eharney has joined #openstack-nova | 13:06 | |
| *** ivasev has joined #openstack-nova | 13:06 | |
| *** toan-tran has left #openstack-nova | 13:07 | |
| *** lifeless has quit IRC | 13:07 | |
| *** joefides has quit IRC | 13:07 | |
| *** joefides has joined #openstack-nova | 13:08 | |
| *** ildikov has joined #openstack-nova | 13:08 | |
| *** adalbas has quit IRC | 13:09 | |
| *** yamahata has joined #openstack-nova | 13:10 | |
| mriedem | ndipanov: have you seen this type of failure before? https://bugs.launchpad.net/ubuntu/+source/nova/+bug/1439280 | 13:12 |
| openstack | Launchpad bug 1439280 in nova (Ubuntu) "Libvirt CPU affinity error" [Undecided,Confirmed] | 13:12 |
| *** mwagner_lap has quit IRC | 13:12 | |
| *** burt has joined #openstack-nova | 13:13 | |
| bauzas | mriedem: hi | 13:13 |
| ndipanov | mriedem, looking | 13:13 |
| bauzas | mriedem: nope, that's unrelated | 13:13 |
| bauzas | mriedem: my patch is fixing the hypervisor related tests | 13:14 |
| bauzas | mriedem: while it leaves 47 failures due to the network issue | 13:14 |
| openstackgerrit | Abhishek Kekane proposed openstack/nova: Improve performance of Unshelve api https://review.openstack.org/150344 | 13:14 |
| bauzas | mriedem: so we can merge either of those independly | 13:14 |
| bauzas | *independently | 13:14 |
| ndipanov | mriedem, not sure about seeing it before but this feature (pinning to a subset of cpus) has been around since icehouse I think | 13:15 |
| bauzas | mriedem: see http://lists.openstack.org/pipermail/openstack-dev/2015-April/061039.html for details | 13:15 |
| ndipanov | it was part of that allowed cpus where you could reserve a subset of cores for non-vm work | 13:15 |
| *** manous has left #openstack-nova | 13:16 | |
| *** vladikr has joined #openstack-nova | 13:16 | |
| ndipanov | when running with tcg - qemu cannot use affinity features maybe (just guessing) | 13:17 |
| ndipanov | since it is single threaded I believe | 13:17 |
| openstackgerrit | sahid proposed openstack/nova: compute: make re-schedule working when resize fails https://review.openstack.org/172044 | 13:17 |
| ndipanov | but that seems too obvious to not be hit for 2 cycle since every ubuntu devstack would be hitting it <- mriedem | 13:18 |
| mriedem | bauzas: yeah i realize they are unrelated, but was thinking if the ducks are in a row and at the end of the tunnet (HEAD patch) we see the cells job as green, it might be easier/more comfortable to approve the changes | 13:19 |
| mriedem | ndipanov: yeah wasn't sure, stephen gordon was saying something about how the code should be working so he wasn't sure why it would be failing on ubuntu | 13:19 |
| mriedem | unless the ubuntu packages are missing something? | 13:20 |
| bauzas | mriedem: sure, we can stack them but it will also depend on a devstack change | 13:20 |
| ndipanov | kashyap, ^^^ | 13:20 |
| ndipanov | any chance you could take a look at the libvirt error in that bug | 13:20 |
| mriedem | bauzas: yeah, but we have people that can approve devstack changes quickly, and it's a small change | 13:21 |
| mriedem | like sdague and mtreinish | 13:21 |
| ndipanov | johnthetubaguy, if you can point me to a wiki that otulines the RPC stuff - I'd be happy to do it... | 13:21 |
| bauzas | mriedem: cool, then let's stack them, any preference for the first ? | 13:22 |
| mriedem | bauzas: well, nvm i guess, since it's a chicken and egg in the nova changes | 13:22 |
| bauzas | mriedem: lol ok | 13:22 |
| ndipanov | https://wiki.openstack.org/wiki/RpcMajorVersionUpdates | 13:23 |
| mriedem | hypervisors will pass with yours, but networks will fail, but in melwitts if you're stacked on top then hypervisors fail o nhers | 13:23 |
| *** neelashah has joined #openstack-nova | 13:23 | |
| mriedem | bauzas: sending yours in | 13:23 |
| johnthetubaguy | ndipanov: cool, I think this is it: https://wiki.openstack.org/wiki/RpcMajorVersionUpdates | 13:23 |
| mriedem | i'll update melwitt's to depend on the devstack change | 13:23 |
| bauzas | mriedem: oh, possible to stack from different projects ? | 13:24 |
| *** atuvenie_ has joined #openstack-nova | 13:24 | |
| ndipanov | cool there are even examples - johnthetubaguy - but this might take a while | 13:24 |
| johnthetubaguy | ndipanov: I was just thinking that :( | 13:25 |
| ndipanov | well we can do it in RC2 if it's ready or not do it... | 13:25 |
| ndipanov | :) | 13:25 |
| johnthetubaguy | ndipanov: yeah, thats very true | 13:27 |
| *** atuvenie has quit IRC | 13:27 | |
| *** jay-lau-513 has joined #openstack-nova | 13:27 | |
| *** rushil has joined #openstack-nova | 13:27 | |
| johnthetubaguy | or we just skip it this time, but feels like its time, since we skipped it last time | 13:27 |
| ndipanov | let me look into this and see waht the impact of not doing this would be | 13:27 |
| ndipanov | i.e. how much code can we actually drop | 13:27 |
| johnthetubaguy | gotcha, thank you! | 13:28 |
| *** dims__ has quit IRC | 13:28 | |
| *** atuvenie__ has joined #openstack-nova | 13:28 | |
| *** ociuhandu_ has joined #openstack-nova | 13:28 | |
| *** dims_ has joined #openstack-nova | 13:28 | |
| bauzas | mriedem: so I'm stacking my patch at the end of the series | 13:29 |
| *** ociuhandu has quit IRC | 13:30 | |
| *** ociuhandu_ is now known as ociuhandu | 13:30 | |
| mriedem | bauzas: hold up | 13:30 |
| mriedem | don't do anything | 13:30 |
| mriedem | i'm on a cal | 13:30 |
| mriedem | call | 13:30 |
| bauzas | mriedem: np | 13:30 |
| alaski | wouldn't we want the devstack change depending on the nova change? | 13:30 |
| mriedem | alaski: the devstack change could live on it's own | 13:31 |
| mriedem | right? | 13:31 |
| *** atuvenie_ has quit IRC | 13:31 | |
| alaski | mriedem: it can, but it's innefective without the nova change | 13:31 |
| mriedem | but melwitt's nova change won't pass the cells job without the devstack change (and bauzas') | 13:31 |
| mriedem | i think my point about stacking originally was that the HEAD patch passes the cells job | 13:31 |
| mriedem | so we can be confident we're not missing something when we merge them all | 13:32 |
| alaski | sure, but HEAD can be the devstack change | 13:32 |
| bauzas | alaski: as I said in the devstack change, I don't see a point to Depend-on | 13:32 |
| *** iamjarvo has joined #openstack-nova | 13:33 | |
| alaski | it doesn't really matter. but conceptually it makes more sense to me that devstack is last | 13:33 |
| mriedem | alaski: does the devstack change run the cells job? | 13:33 |
| *** sudipto has quit IRC | 13:33 | |
| alaski | mriedem: in the experimental queue | 13:33 |
| *** doude has quit IRC | 13:33 | |
| *** dave-mccowan has quit IRC | 13:33 | |
| mriedem | yeah idk, to me the devstack change is 2 lines and it's fine on it's own, it doesn't break anything right? | 13:33 |
| *** dave-mccowan has joined #openstack-nova | 13:34 | |
| mriedem | the nova changes are hairier and i'd want to see those passing with the dependency on the small devstack change | 13:34 |
| mriedem | we could make sure they are passing either way, recheck the devstack change or the nova change, so ultimately doesn't matter probably | 13:34 |
| dansmith | johnthetubaguy: placeholder migrations are the first thing in lemming, not the last thing in kilo | 13:34 |
| alaski | mriedem: what I'm thinking is that the devstack change is essentially a no-op as is, so it's not really tested | 13:35 |
| dansmith | johnthetubaguy: we really need to do an objects bump, but it's too late for that. I had patches up for reviewing the procedure, but they didn't get reviewed | 13:35 |
| alaski | mriedem: basing it on the nova change makes it do something so it's tested | 13:35 |
| *** doude has joined #openstack-nova | 13:35 | |
| alaski | mriedem: but ultimately it's really fine either way | 13:35 |
| dansmith | johnthetubaguy: not sure we really need to do an RPC bump, nor that we have time at this point | 13:35 |
| openstackgerrit | Alex Xu proposed openstack/nova: Add missing default policy rules for servers' actions https://review.openstack.org/166891 | 13:36 |
| *** dims__ has joined #openstack-nova | 13:36 | |
| dansmith | actually, we haven't done one for compute since .. havana? | 13:36 |
| mriedem | alaski: yeah, true, doesn't matter to me i guess | 13:37 |
| dansmith | johnthetubaguy: so if you want I can try to quickly cook up the compute RPC bump. the thing we have to do before rc1 is easier than the thing we do after | 13:37 |
| mriedem | i'm doing two things at once, badly :) | 13:37 |
| *** iamjarvo has quit IRC | 13:37 | |
| *** dims_ has quit IRC | 13:37 | |
| johnthetubaguy | dansmith: hmm, OK, we probably should have tracked a blueprint for this, so they got reviewed, bummer | 13:38 |
| dansmith | johnthetubaguy: maybe so, but alas | 13:38 |
| johnthetubaguy | dansmith: next time | 13:38 |
| dansmith | yeah | 13:38 |
| alaski | mriedem: heh, and I just wanted to be belligerent I guess | 13:39 |
| dansmith | so should we make a stab at the compute bump? | 13:39 |
| dansmith | johnthetubaguy: | 13:39 |
| alaski | mriedem bauzas I'll add a depends-on real quick | 13:39 |
| johnthetubaguy | dansmith: ndipanov said he was going to take a look at how much its worth doing | 13:39 |
| mriedem | alaski: that's fine, i'm taking my anger out on the people bugging me internally :) | 13:39 |
| mriedem | requesting reviews on stuff | 13:39 |
| mriedem | rage! | 13:39 |
| dansmith | johnthetubaguy: I just looked, haven't done it since havana | 13:39 |
| *** rushiagr is now known as rushiagr_away | 13:39 | |
| dansmith | johnthetubaguy: we could dump 40 versions | 13:40 |
| *** dboik has joined #openstack-nova | 13:40 | |
| bauzas | alaski: ok, go for it, I don't want to bikeshed | 13:40 |
| johnthetubaguy | dansmith: that sounds quite attractive, if we can get it in | 13:40 |
| ndipanov | dansmith, that's a lot of code | 13:41 |
| *** exploreshaifali has joined #openstack-nova | 13:41 | |
| ndipanov | if johnthetubaguy is cool with doing it in RC2 | 13:41 |
| dansmith | ndipanov: not to do the bump, it's a lot we get to drop in lemming tho, yeah | 13:41 |
| dansmith | ndipanov: the bump is just calling the 3.40 == 4.0 optionally | 13:41 |
| alaski | dansmith: like run_instance, which I would love to see go away :) | 13:41 |
| dansmith | and a shim class that allows 4.0 | 13:41 |
| kashyap | ndipanov, Sorry, was out for a walk (nice weather here), trying to catch up on the context | 13:41 |
| dansmith | alaski: yeah | 13:41 |
| ndipanov | that hing is still there | 13:41 |
| ndipanov | alaski, ^ | 13:41 |
| ndipanov | yeah I'd say _definitely_worth doing | 13:42 |
| alaski | ndipanov: yep, and totally unused for a while now | 13:42 |
| johnthetubaguy | alaski: +1 | 13:42 |
| ndipanov | so we do only compute then? | 13:42 |
| bauzas | oh oh compute API 4.0 ? I love it | 13:43 |
| kashyap | ndipanov, I don't see a bug URL (for libvirt error) | 13:43 |
| kashyap | ndipanov, Are you referring to the mailing list URL? | 13:43 |
| ndipanov | kashyap, https://bugs.launchpad.net/ubuntu/+source/nova/+bug/1439280 | 13:43 |
| openstack | Launchpad bug 1439280 in nova (Ubuntu) "Libvirt CPU affinity error" [Undecided,Confirmed] | 13:43 |
| * kashyap clicks | 13:44 | |
| * kashyap was off of IRC for 20 minutes, and got to test 4 variations of libvirt migration :-) | 13:44 | |
| kashyap | (Successfully) | 13:44 |
| alaski | mriedem: btw, testing that context/spawn_n change has been a real pain because it relies on code running after a thread is spawned, but I've just about got it I think | 13:44 |
| mriedem | alaski: we can't mock that stuff out? | 13:45 |
| alaski | mriedem: it can be, but not in the just mock it out and assert it was called sort of way | 13:45 |
| dansmith | ndipanov: we only need compute I think | 13:46 |
| dansmith | ndipanov: I'll get going on it here in a few minutes | 13:46 |
| alaski | mriedem: because the thing being mocked needs to actually sort of run | 13:46 |
| mriedem | alaski: blarg! | 13:46 |
| mriedem | so we make a fake or something and stub with that, and do some assertions in the stub? | 13:46 |
| ndipanov | dansmith, many thanks! | 13:46 |
| alaski | mriedem: yeah | 13:46 |
| mriedem | alaski: sounds old school :) | 13:46 |
| alaski | mriedem: and have the stuf run the funtion that's passed in | 13:46 |
| alaski | s/stuf/stub/ | 13:47 |
| *** ganso_ has joined #openstack-nova | 13:47 | |
| kashyap | ndipanov, Your guess above was right "when running with tcg - qemu cannot use affinity features maybe" - I recall learning that from danpb some months ago. | 13:47 |
| alaski | mriedem: heh, like not even using mock at all | 13:47 |
| kashyap | ndipanov, and mriedem: Here's the more technical reason from DanB: | 13:47 |
| kashyap | "It is fundamentally impossible to test CPU pinning with TCG (aka plain QEMU) because TCG only has a single thread for all virtual CPUs. As such there is no mechanism to pin vCPU threads with TCG. Nested KVM is thus the only possible option for testing any of the NUMA / CPU pinning stuff. Instructions for nested KVM setup on a KVM host are documented here" | 13:47 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Honor uuid parameter passed to nova-network create https://review.openstack.org/171911 | 13:47 |
| ndipanov | mriedem, kashyap but waht confuses me is that that feature has been in for some time | 13:48 |
| kashyap | ndipanov, mriedem And, I can confirm that from my testing too. | 13:48 |
| ndipanov | mriedem, let me comment on the bug | 13:48 |
| *** jcoufal has joined #openstack-nova | 13:48 | |
| kashyap | ndipanov, Complete sentences please. "that feature"? | 13:49 |
| ndipanov | mriedem, has context kashyap - sorry | 13:49 |
| ndipanov | allowed_cpus | 13:49 |
| * kashyap looks at the bug and notes this | 13:49 | |
| ndipanov | where you would reserve a subset of CPUs for non-instance use | 13:50 |
| *** tojuvone has quit IRC | 13:50 | |
| mriedem | ndipanov: kashyap: if you guys can update the bug that would be helpful | 13:50 |
| *** tojuvone has joined #openstack-nova | 13:51 | |
| kashyap | mriedem, Yes, trying to look at the bug and refreshing my memory of testing it. And, luckily I have some example XML configs from my testing - https://kashyapc.fedorapeople.org/virt/openstack/nova-libvirt-numa-testing/ | 13:51 |
| kashyap | Don't bother to click, I'll double-check what's in the bug and what's needed. | 13:52 |
| openstackgerrit | sahid proposed openstack/nova-specs: process: add new section ChangeLog https://review.openstack.org/172058 | 13:52 |
| *** openstackgerrit has quit IRC | 13:53 | |
| *** VW_ has joined #openstack-nova | 13:53 | |
| *** VW_ has quit IRC | 13:53 | |
| *** openstackgerrit has joined #openstack-nova | 13:53 | |
| *** VW_ has joined #openstack-nova | 13:54 | |
| *** otter768 has joined #openstack-nova | 13:54 | |
| johnthetubaguy | dansmith: do we do the DB migrations as the first commit of liberty? | 13:55 |
| dansmith | johnthetubaguy: yes | 13:55 |
| *** Nikolay_St has joined #openstack-nova | 13:55 | |
| dansmith | johnthetubaguy: else people on kilo would roll past them immediately and burn them up :) | 13:55 |
| johnthetubaguy | dansmith: yeah, I forgot that bit, oops | 13:56 |
| dansmith | johnthetubaguy: I will get a patch for that up after this and -2 it until we're ready | 13:56 |
| johnthetubaguy | cool, we have the open liberty patch up... | 13:56 |
| dansmith | okay | 13:56 |
| johnthetubaguy | https://review.openstack.org/#/c/171078/ | 13:56 |
| dansmith | I can depends-on that when I do | 13:57 |
| johnthetubaguy | cools, thank you | 13:57 |
| *** mwagner_lap has joined #openstack-nova | 13:58 | |
| *** markvoelker has quit IRC | 13:58 | |
| *** otter768 has quit IRC | 13:59 | |
| *** markvoelker has joined #openstack-nova | 13:59 | |
| sahid | mriedem: quick review :) | 13:59 |
| openstackgerrit | sahid proposed openstack/nova-specs: update spec to refer the change in implemntation https://review.openstack.org/172064 | 14:01 |
| mriedem | sahid: well, it's not an rpm spec | 14:01 |
| mriedem | changelog seems a bit heavy handed | 14:01 |
| sahid | mriedem: it's just a proposition. i had to update the spec virt-driver-large-pages and now operator can not understand why this part or thing has changed | 14:03 |
| sahid | that is why i think is good to have a changelog | 14:03 |
| mriedem | sahid: that'd be useful information in the commit message | 14:03 |
| *** dulek has quit IRC | 14:04 | |
| mriedem | but still, kind of meh | 14:04 |
| sahid | mriedem: well, first they have to understand the spec has changed (since nothing is marked) then they have to know git (operztors are not developers) | 14:05 |
| *** cfriesen has quit IRC | 14:05 | |
| *** cfriesen has joined #openstack-nova | 14:05 | |
| *** browne has quit IRC | 14:05 | |
| *** BadCub_Hiding is now known as BadCub | 14:06 | |
| mriedem | well, as i read it, the proposal is the change log could be read as a thing to update per patch set, which is overkill for a thing that's not merged | 14:06 |
| sahid | we have a good website to print the specs so i think it was good to let users see the history | 14:06 |
| mriedem | sahid: so i'd make that very clear | 14:06 |
| mriedem | it's only for major changes to the spec after it's merged | 14:06 |
| mriedem | and after implementation | 14:06 |
| mriedem | so the only thing in the change log on a spec before it's merged (and before dev starts) is basically * Initial revision. | 14:07 |
| sahid | indeed | 14:07 |
| *** zzzeek has joined #openstack-nova | 14:08 | |
| sahid | mriedem: and it is not so overkill to add a line when something happen on the spec | 14:08 |
| mriedem | sahid: you mean between completely different changes right? | 14:09 |
| mriedem | post-merge of the initial revision | 14:09 |
| *** mtanino has joined #openstack-nova | 14:09 | |
| mriedem | i don't want to see someone putting a change log in patch set 3 that they cleaned up spelling nits from patch set 2 | 14:09 |
| mriedem | that's useless info | 14:10 |
| sahid | ahh no, that is not the point | 14:10 |
| openstackgerrit | Andrey Kurilin proposed openstack/python-novaclient: Adds support for x509 certificates as keypairs https://review.openstack.org/136458 | 14:10 |
| openstackgerrit | Andrey Kurilin proposed openstack/python-novaclient: Implements 'microversions' api type - Part 1 https://review.openstack.org/152569 | 14:10 |
| openstackgerrit | Andrey Kurilin proposed openstack/python-novaclient: Deprecate v1.1 and remove v3 https://review.openstack.org/169378 | 14:10 |
| openstackgerrit | Andrey Kurilin proposed openstack/python-novaclient: Implements 'microversions' api type - Part 2 https://review.openstack.org/167408 | 14:10 |
| *** sdake has joined #openstack-nova | 14:10 | |
| sahid | the point is only when the spec has been merged | 14:10 |
| *** erlon has joined #openstack-nova | 14:10 | |
| bauzas | sahid: you mean when the implemeted diverged most from the spec ? | 14:10 |
| bauzas | *implementation | 14:11 |
| bauzas | or 'is diverging' ? | 14:11 |
| bauzas | sahid: that way, why not just propose a change which would amend the spec ? | 14:11 |
| *** thangp has joined #openstack-nova | 14:11 | |
| bauzas | sahid: we can still blame the file | 14:12 |
| *** jecarey has joined #openstack-nova | 14:13 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: Store context in local store after spawn_n https://review.openstack.org/171299 | 14:13 |
| mriedem | alaski: cool, i'll check that out this afternoon, buried in other stuff this morning | 14:14 |
| mriedem | closing on the sale of our old house :) | 14:14 |
| *** heyongli has quit IRC | 14:14 | |
| alaski | mriedem: nice | 14:14 |
| sahid | bauzas: no sure to understand what you mean, i can understand that there are several ways to find the history of the file yes. | 14:14 |
| *** heyongli has joined #openstack-nova | 14:15 | |
| sahid | the proposal is just to have a simple way for users to follow evolution of the spec | 14:16 |
| *** Jeffrey4l has joined #openstack-nova | 14:17 | |
| *** pcaruana has quit IRC | 14:17 | |
| *** lifeless has joined #openstack-nova | 14:18 | |
| *** annashen has joined #openstack-nova | 14:18 | |
| *** ankit_ag has quit IRC | 14:22 | |
| openstackgerrit | sahid proposed openstack/nova: virt: remove small pages request https://review.openstack.org/172079 | 14:22 |
| bauzas | sahid: mmm, I don't see the point where users need to know the evolution | 14:22 |
| bauzas | sahid: they want to know about the deliverables, not the steps to deliver it | 14:23 |
| *** iamjarvo has joined #openstack-nova | 14:24 | |
| *** iamjarvo has quit IRC | 14:24 | |
| kashyap | mriedem: ndipanov: Commented -- https://bugs.launchpad.net/ubuntu/+source/nova/+bug/1439280 . | 14:25 |
| openstack | Launchpad bug 1439280 in nova (Ubuntu) "Libvirt CPU affinity error" [Undecided,Incomplete] | 14:25 |
| *** iamjarvo has joined #openstack-nova | 14:25 | |
| *** iamjarvo has quit IRC | 14:25 | |
| kashyap | I marked it as "incomplete" for now, since the reporter needs to provide results after testing with nested KVM. | 14:25 |
| ndipanov | kashyap, but that request does not have any numa | 14:25 |
| ndipanov | that's the problem | 14:26 |
| ndipanov | it just uses<vcpu placement='static' cpuset='0-1'>1</vcpu> | 14:26 |
| kashyap | ndipanov: NUMA is not needed, CPU pinning is not possible with TCG | 14:26 |
| *** iamjarvo has joined #openstack-nova | 14:26 | |
| *** jcoufal has quit IRC | 14:26 | |
| kashyap | ndipanov: But, I won't pretend I wrap my head completely around all the corner cases. I myself need to test from current Git this stuff to verify some of the behavior. | 14:27 |
| *** atuvenie__ has quit IRC | 14:27 | |
| *** jcoufal has joined #openstack-nova | 14:27 | |
| *** jcoufal has quit IRC | 14:29 | |
| kashyap | I made a TODO item to create a setup to test it early next week w/ current Nova Git. | 14:29 |
| ndipanov | so kashyap I think your comment is not helpful there because the problem that bug raises is that it fails even if you don't ask for NUMA | 14:29 |
| ndipanov | and you mantion how it's supposed to fail with NUMA, which is not being discussed | 14:29 |
| *** afazekas has quit IRC | 14:29 | |
| *** markus_z has quit IRC | 14:30 | |
| *** rushiagr_away is now known as rushiagr | 14:30 | |
| *** jwcroppe has joined #openstack-nova | 14:31 | |
| openstackgerrit | sahid proposed openstack/nova-specs: process: add new section ChangeLog https://review.openstack.org/172058 | 14:31 |
| kashyap | ndipanov: Hmm, maybe I was too haste in my comment, let me test right away and see what I can come up with. Will correct myself on the bug. | 14:31 |
| *** achanda has joined #openstack-nova | 14:32 | |
| gilliard | mriedem: Hi! Do you have any objection if I put up a new patchset for https://review.openstack.org/#/c/157054 introducing a new config value, defaulting to False? | 14:33 |
| ndipanov | kashyap, the way I see it is this | 14:33 |
| *** sks has quit IRC | 14:34 | |
| * kashyap listens | 14:34 | |
| *** rushiagr is now known as rushiagr_away | 14:34 | |
| openstackgerrit | Chris Gacsal proposed openstack/nova: Update hostname when display_name is changed https://review.openstack.org/170130 | 14:34 |
| mriedem | gilliard: go nuts, i haven't had the time to get back to that recenty | 14:35 |
| ndipanov | kashyap, setting vcpu_pin_set to '' | 14:35 |
| * gilliard goes nuts | 14:35 | |
| ndipanov | should work with GCC | 14:35 |
| *** VW_ has quit IRC | 14:35 | |
| kashyap | ndipanov: TCG you mean | 14:35 |
| ndipanov | setting it to anything but - shouldn't | 14:35 |
| ndipanov | yes | 14:35 |
| ndipanov | TCC | 14:36 |
| *** browne has joined #openstack-nova | 14:36 | |
| mriedem | gilliard: then you'll also need a devstack change on master to set that to True | 14:36 |
| ndipanov | like I said :) | 14:36 |
| mriedem | which depends on hte tempest change | 14:36 |
| openstackgerrit | Chris Gacsal proposed openstack/nova: Update hostname when display_name is changed https://review.openstack.org/170130 | 14:36 |
| gilliard | mriedem: OK that makes sense. The multi-node job is still experimental-only, right? | 14:36 |
| mriedem | gilliard: nope | 14:36 |
| kashyap | ndipanov: A clear test is worth a thousand words, let me test and report the result, so you can tell me where I'm grossly wrong. | 14:36 |
| mriedem | gilliard: it's in check on nova now | 14:36 |
| mriedem | not sure about tempest/devstack | 14:36 |
| mriedem | i think neutron aiocpu is experimental | 14:36 |
| mriedem | nova-network is check | 14:36 |
| * mriedem will be back in a few hours | 14:36 | |
| gilliard | Thanks mriedem | 14:37 |
| *** mriedem has quit IRC | 14:37 | |
| *** annashen has quit IRC | 14:37 | |
| kashyap | ndipanov: Do you still have a moment, so I can describe the problem as I understand it? | 14:37 |
| ndipanov | kashyap, shoot | 14:37 |
| kashyap | ndipanov: Invoking a DevStack instance with <vcpu placement='static' cpuset='0-1'>1</vcpu> | 14:38 |
| *** VW_ has joined #openstack-nova | 14:38 | |
| kashyap | ndipanov: should succeed when the <domain type='qemu'> | 14:38 |
| kashyap | ndipanov: Correct? (Assuming I was comprehensible) | 14:38 |
| ndipanov | kashyap, I think no - it should fail | 14:38 |
| kashyap | (s/DevStack instance/Nova instance) | 14:39 |
| kashyap | ndipanov: Then, it _is_ failing per the reporter? | 14:39 |
| ndipanov | yes | 14:39 |
| *** xuhanp has joined #openstack-nova | 14:39 | |
| *** xuhanp_ has joined #openstack-nova | 14:40 | |
| ndipanov | but that is set by the config option I mentioned above | 14:40 |
| ndipanov | and that config option should be off by default | 14:40 |
| ndipanov | so nova instances should not have that element | 14:40 |
| *** achanda has quit IRC | 14:41 | |
| ndipanov | kashyap, does that make sense?? | 14:41 |
| *** VW__ has joined #openstack-nova | 14:41 | |
| *** salv-orlando has joined #openstack-nova | 14:41 | |
| kashyap | ndipanov: Yes (I hope), I'm reading this commit (by you) from Nova git -- b5df50076239adb9a5a1ef8559d32e46ad9414ae -- libvirt: consider vcpu_pin_set when choosing NUMA cells | 14:42 |
| *** Hosam has joined #openstack-nova | 14:42 | |
| ndipanov | so instances should not have a <vcpu> wotj atr cpuset UNLESS vcpu_pin_set is set | 14:42 |
| *** VW_ has quit IRC | 14:43 | |
| ndipanov | with attr cpuset* | 14:43 |
| ndipanov | that commit is not relevant - but digging into it deeper is difficult because the code moved around | 14:43 |
| kashyap | ndipanov: Okay, got that. Let me test locally (on a slow network) | 14:43 |
| ndipanov | kashyap, cool | 14:43 |
| kashyap | ndipanov: Yes, I was about to say that in this bug, there was no mentoin of NUMA | 14:43 |
| kashyap | ndipanov: I stumbled on that commit as I was grepping for vcpu_pin_set in Nova | 14:43 |
| kashyap | Thanks for your patience :-) | 14:44 |
| *** johnchalekson has joined #openstack-nova | 14:45 | |
| *** ildikov is now known as ildikov_afk | 14:46 | |
| *** VW__ has quit IRC | 14:46 | |
| dansmith | dims__: do you know what oslo.messaging did with the create_rpc_dispatcher function? | 14:46 |
| dansmith | dims__: we used that previously to provide additional_apis for compatibility, but it seems maybe it's gone? | 14:47 |
| dims__ | dansmith: i believe it was never in oslo.messaging | 14:50 |
| dansmith | hu, | 14:50 |
| dansmith | dims__: do you know what (if at all) lets us provide another class to the dispatcher like that supported? | 14:51 |
| dims__ | i pulled stable/juno and stable/icehouse and i don't see it | 14:51 |
| dansmith | dear god | 14:51 |
| dansmith | I hope we're not fscked on that | 14:51 |
| dims__ | dansmith: where is this being used or needs to be used? | 14:52 |
| dansmith | dims__: this is how we support version 3.x and 4.0 at the same time | 14:52 |
| dansmith | dims__: we have two manager classes that each support the different major versions and the dispatcher looks at the message and calls the appropriate one | 14:52 |
| dims__ | last reference i see to that method in nova is https://review.openstack.org/gitweb?p=openstack%2Fnova.git;a=commitdiff;h=d3269fbbbc77dec68b59395c2df251b61d07eb46 | 14:53 |
| dansmith | it's something we've had in nova forever, but the last time we used it was apparently pre-oslo.messaging | 14:53 |
| dansmith | hm, yeah | 14:54 |
| *** salv-orlando has quit IRC | 14:54 | |
| dansmith | seems right | 14:54 |
| dims__ | dansmith: additional_endpoints? | 14:54 |
| dansmith | maybe | 14:54 |
| *** jcoufal has joined #openstack-nova | 14:54 | |
| dansmith | is that in the receive-side dispatcher? | 14:54 |
| dims__ | ConductorManager seems to be using it | 14:54 |
| dims__ | and SchedulerManager | 14:55 |
| dansmith | dims__: conductor is using it for scope separation, not the version thing | 14:55 |
| *** eliqiao has joined #openstack-nova | 14:55 | |
| dansmith | so that might be it, but the dispatcher needs to have the version-awareness for it to work properly | 14:55 |
| dansmith | i.e. it needs to look at the major version of each endpoint and skip the ones where the major doesn't match | 14:55 |
| *** emagana has joined #openstack-nova | 14:55 | |
| *** amotoki has joined #openstack-nova | 14:57 | |
| dims__ | dansmith: ComputeTaskManager and _SchedulerManagerV3Proxy both have target property with a version | 14:57 |
| *** Nikolay_St has quit IRC | 14:57 | |
| dansmith | dims__: right, but they don't expose two different major versions at the same time, right? | 14:58 |
| dansmith | oh, sched proxy | 14:58 |
| dansmith | yes, that's it | 14:58 |
| dansmith | sched proxy should be gone by now (!) | 14:58 |
| dansmith | but yeah, awesome | 14:58 |
| dansmith | dims__: thanks! | 14:58 |
| dims__ | :) | 14:58 |
| *** baoli has quit IRC | 15:00 | |
| *** dboik_ has joined #openstack-nova | 15:00 | |
| *** johnchalekson has quit IRC | 15:00 | |
| *** baoli has joined #openstack-nova | 15:01 | |
| *** marun has joined #openstack-nova | 15:02 | |
| *** artom has joined #openstack-nova | 15:02 | |
| *** xuhanp has quit IRC | 15:02 | |
| *** xuhanp_ has quit IRC | 15:02 | |
| *** achanda has joined #openstack-nova | 15:02 | |
| *** dboik has quit IRC | 15:04 | |
| *** kylek3h has quit IRC | 15:04 | |
| *** jcoufal has quit IRC | 15:06 | |
| *** tonytan4ever has joined #openstack-nova | 15:07 | |
| *** Mike_D_laptop has joined #openstack-nova | 15:07 | |
| *** dsanders has joined #openstack-nova | 15:09 | |
| *** yamahata has quit IRC | 15:10 | |
| *** achanda has quit IRC | 15:11 | |
| *** gszasz has quit IRC | 15:11 | |
| *** tojuvone has quit IRC | 15:12 | |
| *** jistr is now known as jistr|mtg | 15:15 | |
| eliqiao | hi all, I meet this issue, how can I fix it ? in cinder , it requirest oslo.config>=1.9.3,<1.10.0 # Apache-2.0 | 15:16 |
| eliqiao | but in keystone it require oslo.config>=1.4.0,<=1.6.0 # Apache-2.0 | 15:16 |
| dansmith | alaski: run_instance() is still in our rpcapi | 15:16 |
| dansmith | alaski: presumably I can snip that out now, right? | 15:16 |
| *** annashen has joined #openstack-nova | 15:16 | |
| *** annashen has quit IRC | 15:17 | |
| alaski | dansmith: yes. the only caller was the scheduler and that was removed when it was bumped to 4.x | 15:19 |
| dansmith | okay sweet | 15:19 |
| * dansmith puts a 9mm hole in it | 15:19 | |
| eliqiao | ahh.. can any one give me a hand :( | 15:20 |
| kashyap | ndipanov: Okay, seems like I'm able to reproduce the issue, w/ current Nova git (and very latest libvirt/QEMU), see line 75 -- http://paste.openstack.org/show/201358/ | 15:20 |
| kashyap | ndipanov: So, if IIUC, that 'cpuset' attribute should not be there, *unless*, I set it explicitly to that flavor: 'nova flavor-key m1.tiny set hw:vcpu_pin_set=0-1' | 15:22 |
| dims__ | eliqiao: trunk versions of both projects? | 15:22 |
| *** armax has joined #openstack-nova | 15:22 | |
| dims__ | eliqiao: both project should have the same versions of stuff from https://github.com/openstack/requirements/blob/master/global-requirements.txt | 15:23 |
| ndipanov | no | 15:23 |
| ndipanov | unless you set the config option on the compute node | 15:23 |
| dims__ | eliqiao: looks like you have an old keystone? https://github.com/openstack/keystone/blob/master/requirements.txt#L22 | 15:24 |
| eliqiao | dims_: thx ,my stupid | 15:24 |
| eliqiao | dims_: I don't aware that the requirement changed by devstack or ... | 15:25 |
| dims__ | eliqiao: i throw in RECLONE=yes in localrc, that pulls latest from git everytime | 15:26 |
| *** ajayaa has quit IRC | 15:26 | |
| eliqiao | dims_: sure, but I need use juno version ... | 15:26 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Fix for live migration rollback when using shared storage https://review.openstack.org/172117 | 15:27 |
| eliqiao | dims_: I reset cinder repo and will try devstack again. | 15:27 |
| *** annashen has joined #openstack-nova | 15:27 | |
| dims__ | eliqiao: ack | 15:27 |
| *** jistr|mtg is now known as jistr | 15:27 | |
| kashyap | ndipanov: Ignorant question: set on the Compute node's nova.conf like that? vcpu_pin_set=0-1 | 15:27 |
| *** annashen has quit IRC | 15:28 | |
| kashyap | ndipanov: And, looking at that XML snippet from that pastebin, does that look like a bug? | 15:28 |
| *** johnchalekson has joined #openstack-nova | 15:28 | |
| *** ildikov_afk has quit IRC | 15:29 | |
| ndipanov | kashyap, as a string | 15:29 |
| ndipanov | so ='0-3' | 15:29 |
| ndipanov | but it should not be there by default | 15:29 |
| ndipanov | can you comment on the bug with your findings, and I can try to figure out why it sets it even if it's not set up | 15:30 |
| kashyap | ndipanov: It's not there by default. | 15:30 |
| kashyap | ndipanov: Yes, I'm going to correct myself on the bug noting this detail. | 15:30 |
| *** dboik_ has quit IRC | 15:30 | |
| *** dsanders has quit IRC | 15:31 | |
| *** joefides has quit IRC | 15:31 | |
| kashyap | (I mean, "not there by default" in the config file at-least, I'm using a AIO DevStack.) | 15:31 |
| *** dboik has joined #openstack-nova | 15:31 | |
| *** tsekiyama has joined #openstack-nova | 15:31 | |
| ndipanov | kashyap, yeah - it should not be set by default | 15:32 |
| ndipanov | kashyap, did you run without virt=kvm | 15:33 |
| ndipanov | and actually see it fail? | 15:33 |
| kashyap | ndipanov: Yes (w/o KVM) | 15:33 |
| *** tsekiyama has quit IRC | 15:33 | |
| kashyap | ndipanov: Hang on, I misspoke, let me correct myself. | 15:33 |
| kashyap | ndipanov: With my test: (a) The instance boots successfully. (b) But, as we saw, the 'cpuset' attribute was set, with QEMU domain | 15:34 |
| kashyap | (I actually unloaded the KVM kernel module before running DevStack), I'll attach my nova.conf too. | 15:34 |
| ndipanov | kashyap, so then that bug is not with the nova configuration | 15:35 |
| kashyap | ndipanov: So, when I said "I was able to reproduce" -- I was talking only about the 'cpuset' attribute. | 15:35 |
| ndipanov | just attach your findings | 15:35 |
| kashyap | Okido. | 15:35 |
| ndipanov | and if your qemu does not barf on it | 15:35 |
| ndipanov | which it doesn't | 15:36 |
| ndipanov | let them figure out what they did worng | 15:36 |
| kashyap | No, it doesn't. I double-ensured by checking the QEMU CLI that was invoked. | 15:36 |
| kashyap | Yep | 15:36 |
| ndipanov | kashyap, paste that too | 15:36 |
| ndipanov | awesome! | 15:36 |
| ndipanov | NOP | 15:36 |
| bauzas | dims__: hadn't time to say congrats btw. for your PTL position | 15:36 |
| kashyap | Yep, was planning to do anyway. Completeness for win! | 15:36 |
| *** salv-orlando has joined #openstack-nova | 15:36 | |
| kashyap | ndipanov: Is that really "awesome"? :-) | 15:37 |
| *** rushil has quit IRC | 15:37 | |
| *** atuvenie__ has joined #openstack-nova | 15:38 | |
| *** dave-mccowan has quit IRC | 15:38 | |
| *** jistr has quit IRC | 15:40 | |
| *** mriedem has joined #openstack-nova | 15:47 | |
| *** jay-lau-513 has quit IRC | 15:47 | |
| *** oro has quit IRC | 15:47 | |
| *** jay-lau-513 has joined #openstack-nova | 15:47 | |
| *** arnaud___ has joined #openstack-nova | 15:47 | |
| *** adalbas has joined #openstack-nova | 15:49 | |
| *** Marga_ has joined #openstack-nova | 15:53 | |
| *** Marga_ has quit IRC | 15:53 | |
| jogo | rajesht: pong | 15:54 |
| *** Marga_ has joined #openstack-nova | 15:54 | |
| *** otter768 has joined #openstack-nova | 15:55 | |
| *** dave-mccowan has joined #openstack-nova | 15:56 | |
| *** otter768 has quit IRC | 16:00 | |
| *** yamahata has joined #openstack-nova | 16:02 | |
| *** rwsu has joined #openstack-nova | 16:02 | |
| kashyap | ndipanov: Corrected myself, and noted it -- https://bugs.launchpad.net/ubuntu/+source/nova/+bug/1439280/comments/5 | 16:04 |
| openstack | Launchpad bug 1439280 in nova (Ubuntu) "Libvirt CPU affinity error" [Undecided,Incomplete] | 16:04 |
| *** Hosam has quit IRC | 16:04 | |
| ndipanov | kashyap, awesome many thanks | 16:05 |
| *** rushil has joined #openstack-nova | 16:05 | |
| *** rushil has quit IRC | 16:06 | |
| *** jlanoux has quit IRC | 16:06 | |
| *** atuvenie__ has quit IRC | 16:07 | |
| kashyap | ndipanov: Thanks for pushing me and to lazily defer it to next week :-) | 16:09 |
| kashyap | and *not to | 16:09 |
| ndipanov | kashyap, anything for ubuntu friends :) | 16:09 |
| *** derekh has quit IRC | 16:12 | |
| *** eliqiao has quit IRC | 16:13 | |
| *** hemnafk is now known as hemna | 16:16 | |
| *** lucasagomes has quit IRC | 16:16 | |
| *** signed8bit has joined #openstack-nova | 16:18 | |
| openstackgerrit | Chris Friesen proposed openstack/nova: libvirt: check qemu version for NUMA & hugepage support https://review.openstack.org/170780 | 16:19 |
| *** rajesht has quit IRC | 16:19 | |
| *** tjones1 has joined #openstack-nova | 16:19 | |
| *** vilobhmm1 has joined #openstack-nova | 16:20 | |
| *** tjones1 has left #openstack-nova | 16:20 | |
| *** ndipanov has quit IRC | 16:20 | |
| *** ajayaa has joined #openstack-nova | 16:21 | |
| *** vilobhmm11 has joined #openstack-nova | 16:23 | |
| *** apuimedo has quit IRC | 16:24 | |
| *** ozialien has quit IRC | 16:26 | |
| *** romainh has left #openstack-nova | 16:26 | |
| *** melwitt has joined #openstack-nova | 16:26 | |
| *** vilobhmm1 has quit IRC | 16:27 | |
| *** annashen has joined #openstack-nova | 16:27 | |
| *** arnaud___ has quit IRC | 16:29 | |
| *** dguitarbite has quit IRC | 16:34 | |
| *** Nic has joined #openstack-nova | 16:34 | |
| *** exploreshaifali has quit IRC | 16:34 | |
| *** foexle_ has quit IRC | 16:34 | |
| *** harlowja_away is now known as harlowja | 16:35 | |
| *** signed8b_ has joined #openstack-nova | 16:37 | |
| *** rushil has joined #openstack-nova | 16:39 | |
| *** signed8b_ has quit IRC | 16:40 | |
| *** signed8bit has quit IRC | 16:40 | |
| melwitt | alaski, mriedem: is it worth moving the Depends-On to the devstack patch? or did you all decide it made more sense on the nova one? | 16:40 |
| mriedem | melwitt: i think we decided that it didn't matter ultimately | 16:41 |
| *** signed8bit has joined #openstack-nova | 16:41 | |
| *** ijw has joined #openstack-nova | 16:41 | |
| *** jichen has quit IRC | 16:41 | |
| *** patrickeast has joined #openstack-nova | 16:42 | |
| *** tonytan4ever has quit IRC | 16:42 | |
| melwitt | mriedem: okay :) I'm adding the Related-Bug tag to the devstack anyway so wanted to check | 16:42 |
| mriedem | can we have circular deps with depends-on? | 16:42 |
| mriedem | mtreinish: sdague: ^? | 16:42 |
| *** Mike_D_laptop has quit IRC | 16:42 | |
| mriedem | i think we can | 16:42 |
| mtreinish | mriedem: no | 16:42 |
| mriedem | oh | 16:42 |
| mriedem | drats | 16:42 |
| *** signed8bit has quit IRC | 16:43 | |
| mtreinish | mriedem: http://lists.openstack.org/pipermail/openstack-dev/2015-February/056515.html | 16:43 |
| melwitt | well, I would change the other one too, to remove it from the nova | 16:43 |
| melwitt | first I guess. would have to be | 16:43 |
| *** signed8bit has joined #openstack-nova | 16:43 | |
| mtreinish | mriedem: the second bullet | 16:43 |
| sdague | mriedem: if it detects a cycle it just punts on them all | 16:43 |
| dansmith | mriedem: I'm going to want you to look at this rpc version bump patch in a bit | 16:43 |
| mriedem | oh heavens! | 16:44 |
| mriedem | i'm about to rage quit internally, so ok | 16:44 |
| melwitt | hells bells | 16:44 |
| mriedem | martha | 16:44 |
| *** rushil has quit IRC | 16:45 | |
| *** rushil has joined #openstack-nova | 16:45 | |
| *** signed8bit has quit IRC | 16:45 | |
| *** signed8bit has joined #openstack-nova | 16:46 | |
| * mriedem is about to jump in the swift irc channel and get ripped apart | 16:46 | |
| sdague | mriedem: sounds like fun | 16:47 |
| mriedem | well, someone has to do it | 16:47 |
| mriedem | given i'm the swift sme today, i guess it has to be me | 16:47 |
| *** dsanders has joined #openstack-nova | 16:48 | |
| dansmith | I would almost rather do that than this rpc bump | 16:49 |
| dansmith | ...almost | 16:49 |
| *** signed8bit has quit IRC | 16:50 | |
| bauzas | melwitt: so thanks to mriedem, we now only have the netwoirk failures, no longer the hyp ones | 16:50 |
| *** signed8bit has joined #openstack-nova | 16:50 | |
| *** oro has joined #openstack-nova | 16:51 | |
| bauzas | melwitt: so if you want to do depends-on, let's do the devstack one depending on the last nova bugfix | 16:51 |
| bauzas | melwitt: and run the experimental pipeline | 16:51 |
| melwitt | bauzas: I saw :) I'm glad | 16:51 |
| *** signed8bit has quit IRC | 16:52 | |
| *** signed8bit has joined #openstack-nova | 16:52 | |
| melwitt | bauzas: yes, in the middle of doing that. wanted to fixup the unit test with your comment. looking at it, I would have to anyway assign the result of create_networks to a variable in order to check the content. and I thought it would make sense to do like "assertTrue(nets)" rather than assertEqual(1, len(nets)). did you have something in mind? | 16:53 |
| *** unicell has joined #openstack-nova | 16:53 | |
| melwitt | s/would/wouldn't/ | 16:53 |
| mriedem | sdague: dansmith: i'm actually pleasantly surprised by helpfulness right now :) | 16:54 |
| *** signed8bit has quit IRC | 16:54 | |
| *** signed8bit has joined #openstack-nova | 16:54 | |
| *** david8hu has quit IRC | 16:54 | |
| *** Nic has quit IRC | 16:54 | |
| *** tpatil has joined #openstack-nova | 16:54 | |
| *** Jeffrey4l has quit IRC | 16:55 | |
| *** dsanders1 has joined #openstack-nova | 16:55 | |
| *** tjones1 has joined #openstack-nova | 16:56 | |
| *** aswadr has quit IRC | 16:57 | |
| *** burt has quit IRC | 16:57 | |
| openstackgerrit | melanie witt proposed openstack/nova: Honor uuid parameter passed to nova-network create https://review.openstack.org/171911 | 16:58 |
| *** dsanders has quit IRC | 16:58 | |
| *** Nic has joined #openstack-nova | 16:59 | |
| *** VW_ has joined #openstack-nova | 16:59 | |
| *** sahid has quit IRC | 17:01 | |
| *** signed8bit has quit IRC | 17:03 | |
| *** lpetrut has left #openstack-nova | 17:03 | |
| *** signed8bit has joined #openstack-nova | 17:03 | |
| *** smccully has joined #openstack-nova | 17:05 | |
| *** isd has joined #openstack-nova | 17:06 | |
| *** signed8bit has quit IRC | 17:13 | |
| *** signed8bit has joined #openstack-nova | 17:13 | |
| *** openstack has quit IRC | 17:13 | |
| *** openstack has joined #openstack-nova | 17:14 | |
| *** exploreshaifali has joined #openstack-nova | 17:15 | |
| *** signed8bit has quit IRC | 17:15 | |
| *** thomasem has quit IRC | 17:15 | |
| *** signed8bit has joined #openstack-nova | 17:16 | |
| *** achanda has joined #openstack-nova | 17:17 | |
| bauzas | melwitt: my comment was about saying that you just added a new conditional, so it was not needed to change the existing test_create_networks() | 17:18 |
| bauzas | melwitt: I would have preferred that you leave test_create_networks() and adding a new test_create_networks_with_uuid() method like the one you added in PS4, and checking net[uuid] | 17:19 |
| melwitt | bauzas: oh, make an entirely new test. sorry, I misunderstood | 17:19 |
| bauzas | melwitt: yeah, because if not passing an uuid, you're not checking that it still works | 17:19 |
| *** baoli has quit IRC | 17:19 | |
| bauzas | melwitt: well, tbh, it's covered by other tests but I think it was clearer | 17:20 |
| *** doude has quit IRC | 17:20 | |
| melwitt | bauzas: gotcha | 17:20 |
| *** Marga_ has quit IRC | 17:21 | |
| *** amotoki has quit IRC | 17:21 | |
| *** doude has joined #openstack-nova | 17:21 | |
| *** sbfox has joined #openstack-nova | 17:23 | |
| *** Marga_ has joined #openstack-nova | 17:24 | |
| *** salv-orlando has quit IRC | 17:24 | |
| *** eglynn has quit IRC | 17:25 | |
| openstackgerrit | melanie witt proposed openstack/nova: Honor uuid parameter passed to nova-network create https://review.openstack.org/171911 | 17:25 |
| openstackgerrit | Dan Smith proposed openstack/nova: Update compute version alias for juno https://review.openstack.org/172151 | 17:25 |
| openstackgerrit | Dan Smith proposed openstack/nova: Add compute RPC API v4.0 https://review.openstack.org/172152 | 17:25 |
| dansmith | mriedem: ^ | 17:25 |
| dansmith | and others | 17:25 |
| dansmith | we really need that | 17:26 |
| *** ozialien has joined #openstack-nova | 17:26 | |
| dansmith | it's almost completely mechanical | 17:26 |
| dansmith | we have to do it right before the rc, and haven't since havana :( | 17:26 |
| *** neelashah has joined #openstack-nova | 17:27 | |
| *** boris-42 has quit IRC | 17:28 | |
| *** neelashah1 has joined #openstack-nova | 17:28 | |
| mriedem | aight | 17:28 |
| mriedem | gotta wrap up my swift endeavour | 17:28 |
| *** zzzeek has quit IRC | 17:29 | |
| *** baoli has joined #openstack-nova | 17:29 | |
| *** Marga_ has quit IRC | 17:29 | |
| *** lpetrut has joined #openstack-nova | 17:30 | |
| dansmith | I'm adding comments to help reviewers | 17:31 |
| *** dboik has quit IRC | 17:31 | |
| *** neelashah has quit IRC | 17:32 | |
| *** tonytan4ever has joined #openstack-nova | 17:35 | |
| mriedem | alaski: not easy enough! https://review.openstack.org/#/c/172151/ | 17:36 |
| mriedem | :) | 17:36 |
| *** tonytan4ever has quit IRC | 17:36 | |
| alaski | mriedem: argh | 17:36 |
| alaski | fair point though | 17:37 |
| *** aix has quit IRC | 17:37 | |
| alaski | there should probably just be one generic comment above VERSION_ALIASES | 17:37 |
| *** emagana has quit IRC | 17:37 | |
| dims__ | dansmith: slightly confusing to see that we are adding kilo alias and commit message talks about juno :) | 17:38 |
| *** emagana has joined #openstack-nova | 17:38 | |
| dansmith | alaski: mriedem: that comment is in the next patch | 17:38 |
| *** zzzeek has joined #openstack-nova | 17:38 | |
| dansmith | dims__: why? (doh) | 17:38 |
| dansmith | :P | 17:38 |
| alaski | woo, vindicated | 17:38 |
| mriedem | gilliard: https://review.openstack.org/#/c/164427/ | 17:39 |
| mriedem | that's your reference devstack change for the tempest thingy | 17:39 |
| dims__ | :) | 17:39 |
| dansmith | dims__: slap a -1 on there, I don't want to overwrite my comments on the second patch just yet, until people can have a look | 17:39 |
| *** tonytan4ever has joined #openstack-nova | 17:41 | |
| dansmith | wtf, I pass functional tests locally | 17:42 |
| * dansmith makes a hand gesture at jenkins | 17:42 | |
| *** tjones1 has quit IRC | 17:42 | |
| *** emagana has quit IRC | 17:43 | |
| *** tjones1 has joined #openstack-nova | 17:43 | |
| *** david8hu has joined #openstack-nova | 17:46 | |
| *** dboik has joined #openstack-nova | 17:46 | |
| mriedem | alaski: tempest/devstack was very unhappy with your change :( https://review.openstack.org/#/c/171299/ | 17:46 |
| alaski | mriedem: yeah, it's an issue with getcallargs. it apparently doesn't work when a method signature just has (*args, **kwargs) | 17:46 |
| alaski | working on a fix now | 17:47 |
| mriedem | ah that sucks | 17:47 |
| *** signed8bit has quit IRC | 17:47 | |
| openstackgerrit | melanie witt proposed openstack/nova: Refactor nova-net cidr validation in prep for bug fix https://review.openstack.org/171912 | 17:48 |
| openstackgerrit | melanie witt proposed openstack/nova: Honor uuid parameter passed to nova-network create https://review.openstack.org/171911 | 17:48 |
| *** salv-orlando has joined #openstack-nova | 17:51 | |
| *** thomasem has joined #openstack-nova | 17:52 | |
| *** Sukhdev has joined #openstack-nova | 17:52 | |
| *** thomasem has joined #openstack-nova | 17:53 | |
| dansmith | doh, forgot to hit save on the review notes | 17:53 |
| dansmith | up there now | 17:53 |
| *** VW_ has quit IRC | 17:53 | |
| *** moshele has joined #openstack-nova | 17:53 | |
| *** ildikov_afk has joined #openstack-nova | 17:55 | |
| *** Marga_ has joined #openstack-nova | 17:56 | |
| *** otter768 has joined #openstack-nova | 17:56 | |
| *** VW_ has joined #openstack-nova | 17:58 | |
| *** baoli has quit IRC | 17:59 | |
| openstackgerrit | Dan Smith proposed openstack/nova: Add compute RPC API v4.0 https://review.openstack.org/172152 | 17:59 |
| openstackgerrit | Dan Smith proposed openstack/nova: Update compute version alias for kilo https://review.openstack.org/172151 | 17:59 |
| *** baoli has joined #openstack-nova | 17:59 | |
| *** ZZelle_ has joined #openstack-nova | 18:00 | |
| mriedem | vishy: there isn't any wonky reason why nova-network wouldn't honor the uuid passed in when creating networks is there? | 18:00 |
| mriedem | https://review.openstack.org/#/c/171911/ | 18:00 |
| *** otter768 has quit IRC | 18:01 | |
| *** baoli has quit IRC | 18:01 | |
| *** doude has quit IRC | 18:05 | |
| *** patrickeast_ has joined #openstack-nova | 18:05 | |
| *** IanGovett has quit IRC | 18:05 | |
| *** sdake_ has joined #openstack-nova | 18:05 | |
| *** doude has joined #openstack-nova | 18:06 | |
| *** patrickeast has quit IRC | 18:06 | |
| *** patrickeast_ is now known as patrickeast | 18:06 | |
| *** Marga_ has quit IRC | 18:07 | |
| *** mestery has quit IRC | 18:07 | |
| *** patrickeast has quit IRC | 18:08 | |
| *** dulek has joined #openstack-nova | 18:08 | |
| *** VW_ has quit IRC | 18:09 | |
| *** sdake has quit IRC | 18:09 | |
| mriedem | dansmith: 2 comments in ps1 of https://review.openstack.org/#/c/172152/ | 18:10 |
| *** mestery has joined #openstack-nova | 18:10 | |
| mriedem | melwitt: smallish thing in here and then i'm +2 https://review.openstack.org/#/c/171912/ | 18:11 |
| *** sbfox has quit IRC | 18:11 | |
| melwitt | mriedem: already on it | 18:11 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Store context in local store after spawn_n https://review.openstack.org/171299 | 18:12 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Avoid indexing into an empty list in getcallargs https://review.openstack.org/172163 | 18:12 |
| alaski | mriedem: that should make things better | 18:12 |
| *** VW_ has joined #openstack-nova | 18:12 | |
| *** VW_ has quit IRC | 18:12 | |
| *** VW_ has joined #openstack-nova | 18:13 | |
| mriedem | alaski: nice, just waiting on jenkins | 18:14 |
| openstackgerrit | melanie witt proposed openstack/nova: Refactor nova-net cidr validation in prep for bug fix https://review.openstack.org/171912 | 18:15 |
| openstackgerrit | melanie witt proposed openstack/nova: Honor uuid parameter passed to nova-network create https://review.openstack.org/171911 | 18:15 |
| *** rajesht has joined #openstack-nova | 18:15 | |
| *** Marga_ has joined #openstack-nova | 18:18 | |
| *** moshele has quit IRC | 18:19 | |
| *** rajesht has quit IRC | 18:19 | |
| *** Marga_ has quit IRC | 18:20 | |
| *** Marga_ has joined #openstack-nova | 18:20 | |
| *** sdake has joined #openstack-nova | 18:25 | |
| *** vilobhmm11 has quit IRC | 18:29 | |
| *** sdake_ has quit IRC | 18:29 | |
| *** ajayaa has quit IRC | 18:30 | |
| *** vilobhmm1 has joined #openstack-nova | 18:30 | |
| artom | Just making sure, but in the official API docs (http://developer.openstack.org/api-ref-compute-v2.1.html), 'server' == 'instance', yes? | 18:31 |
| *** sbfox has joined #openstack-nova | 18:32 | |
| *** ijw has quit IRC | 18:33 | |
| melwitt | artom: yes | 18:34 |
| *** exploreshaifali has quit IRC | 18:35 | |
| *** mspreitz has joined #openstack-nova | 18:37 | |
| *** clarkb has joined #openstack-nova | 18:38 | |
| clarkb | I just discovered a bug with shelve working in the multinode devstack gate test envs. I have two nodes a controller with compute node and a compute node. I boot two VMs test1 and test2 and attach fip1 to test1 and fip2 to test2. Then I delete test2, shelve test1, unshelve test1 | 18:39 |
| clarkb | with the shelve then unshelve test1 moves from controller to compute node | 18:40 |
| clarkb | test1 gets test2's old dhcp address | 18:40 |
| clarkb | the funny thing is now test1 is attached to both fip1 and fip2 because fip2 wasn't cleaned up when test2 was deleted | 18:40 |
| clarkb | is this a known bug? if not I will go file it, but wanted to check before I dug around in nova's massive bug tracker | 18:41 |
| artom | melwitt, thanks! | 18:41 |
| *** ozialien has quit IRC | 18:42 | |
| *** jichen has joined #openstack-nova | 18:44 | |
| *** ozialien has joined #openstack-nova | 18:44 | |
| *** zzzeek has quit IRC | 18:46 | |
| *** sbfox has quit IRC | 18:48 | |
| *** doude has quit IRC | 18:51 | |
| dansmith | alaski: hmm, can't I just remove run_instance() from manager? I forgot to do it | 18:52 |
| alaski | dansmith: I think so, because it can't be used anymore. but doesn't it need to stay for compatibility? | 18:53 |
| dansmith | alaski: oh, right, it does until I actually remove the 3.x manager, yeah | 18:54 |
| dansmith | alaski: so we can remove it in lemming | 18:54 |
| dansmith | mriedem: alaski: thanks a lot for jumping on this | 18:54 |
| alaski | dansmith: cool. it's sort of an odd case because it was only called from the scheduler which was already removed. so though the rpc is still there nothing has called it for a release | 18:55 |
| *** aix has joined #openstack-nova | 18:55 | |
| mriedem | clarkb: nova-network? | 18:56 |
| dansmith | yeah | 18:56 |
| dansmith | that's fine, it will be gone soon enough :) | 18:56 |
| mriedem | clarkb: i don't know the answer about the bug, but sounds like a bug - might as well throw another on the pile | 18:56 |
| *** jecarey has quit IRC | 18:57 | |
| vilobhmm1 | proposed a spec https://review.openstack.org/#/c/138607/ under https://review.openstack.org/#/c/138607/10/specs/liberty/approved/service-group-using-tooz.rst is "specs/liberty/approved" right place for new specs to be put for review ? | 18:58 |
| vilobhmm1 | dansmith, mriedem, melwitt : ^^^ | 18:59 |
| mriedem | vilobhmm1: yes | 19:01 |
| mriedem | even though they aren't approved yet, which is confusing | 19:01 |
| vilobhmm1 | mriedem : yep that was the confusing part as it two dir "approved" and "implemented" | 19:01 |
| vilobhmm1 | thanks mriedem | 19:02 |
| dansmith | you're proposing it to be approved | 19:02 |
| mriedem | process of elimination :) | 19:02 |
| vilobhmm1 | dansmith : sure :) | 19:02 |
| *** annashen has quit IRC | 19:03 | |
| *** VW_ has quit IRC | 19:03 | |
| *** Marga_ has quit IRC | 19:03 | |
| *** sbfox has joined #openstack-nova | 19:04 | |
| *** jecarey has joined #openstack-nova | 19:04 | |
| vilobhmm1 | dansmith, mriedem : since this spec was proposed for kilo ; i need to re-submit it for liberty under the specs/liberty/approved. sounds good ? | 19:04 |
| dansmith | yes | 19:04 |
| vilobhmm1 | cool | 19:04 |
| vilobhmm1 | thnx | 19:05 |
| bauzas | dansmith: I left a note on your patch about inversing the logic by providing a V3Proxy | 19:05 |
| *** Marga_ has joined #openstack-nova | 19:05 | |
| bauzas | dansmith: that way, it could maybe be better understandable what is deprecated and what's not | 19:05 |
| dansmith | bauzas: doing it this way is how we've done it in the past, | 19:06 |
| bauzas | but that's MHO and I know it's not the usual way | 19:06 |
| bauzas | eh, was just typing that | 19:06 |
| dansmith | and since it takes a while to wrap your head around it each time, consistency is good | 19:06 |
| bauzas | dansmith: not exactly, see the Scheduler RPC API | 19:06 |
| *** Marga_ has quit IRC | 19:06 | |
| dansmith | I think this is also less work right now, which is when it counts | 19:06 |
| dansmith | bauzas: see the last rev of the compute rpc api | 19:06 |
| bauzas | dansmith: but the Sched RPC API is having only 2 methods | 19:06 |
| *** Marga_ has joined #openstack-nova | 19:06 | |
| dansmith | yeah | 19:07 |
| bauzas | dansmith: yeah I know | 19:07 |
| bauzas | dansmith: I left a +1 anyway but I was wondering if it was maybe not better the other way | 19:07 |
| *** mmedvede has quit IRC | 19:07 | |
| dansmith | if we do the subclass and the v3 proxy, then we have to remember not to allow any changes to the main v4 class, | 19:09 |
| dansmith | else they will be reachable via the v3 subclass | 19:09 |
| dansmith | which would be bad | 19:09 |
| dansmith | this is a little heavy, but is very explicit | 19:09 |
| dansmith | damn, I failed cells | 19:10 |
| dansmith | alaski: ^ | 19:10 |
| alaski | dansmith: looking | 19:12 |
| dansmith | we can't see the deets yet | 19:12 |
| *** VW_ has joined #openstack-nova | 19:12 | |
| *** IanGovett has joined #openstack-nova | 19:13 | |
| dansmith | I'm super thrilled that that all passed tempest | 19:13 |
| *** dulek has quit IRC | 19:14 | |
| alaski | yeah, all I could get at was https://jenkins01.openstack.org/job/check-devstack-dsvm-cells/19505/console which doesn't give much info | 19:15 |
| dansmith | it's there now | 19:17 |
| dansmith | alaski: http://logs.openstack.org/52/172152/2/check/check-devstack-dsvm-cells/2f7a13e/logs/screen-n-cpu.txt.gz?level=TRACE | 19:18 |
| mriedem | dansmith: you're going to fail cells | 19:18 |
| dansmith | ugh | 19:18 |
| mriedem | let's see if that's just you | 19:19 |
| mriedem | heh, it is :) | 19:19 |
| alaski | dansmith: only one, that's not so bad | 19:19 |
| alaski | dansmith: the real solution is https://review.openstack.org/#/c/79741/2/nova/cells/messaging.py,cm | 19:20 |
| *** dprince has quit IRC | 19:20 | |
| alaski | but adding detach_volume to EXPECTS_OBJECTS works as well | 19:20 |
| dansmith | alaski: why is this only cropping up on this patch though? | 19:20 |
| dansmith | I'm not changing anything that should care, right? | 19:20 |
| alaski | yeah, not that I can see | 19:21 |
| *** vilobhmm1 has quit IRC | 19:21 | |
| dansmith | oh I got it | 19:22 |
| *** Longgeek_ has quit IRC | 19:22 | |
| *** vilobhmm1 has joined #openstack-nova | 19:22 | |
| dansmith | detach_volume has its arguments reversed | 19:22 |
| mriedem | yup | 19:22 |
| dansmith | although | 19:23 |
| dansmith | not sure that should matter | 19:23 |
| *** ijw has joined #openstack-nova | 19:23 | |
| mriedem | b/c of kwargs? | 19:23 |
| openstackgerrit | Dan Smith proposed openstack/nova: Add compute RPC API v4.0 https://review.openstack.org/172152 | 19:23 |
| openstackgerrit | Clif Houck proposed openstack/nova: Add weigher for cached images on Ironic nodes. https://review.openstack.org/170681 | 19:25 |
| dansmith | right | 19:25 |
| dansmith | because the thing that was switched should be dispatched by kwarg | 19:26 |
| *** iamjarvo has quit IRC | 19:26 | |
| dansmith | and it works for the normal case | 19:26 |
| dansmith | but maybe the dispatching happens differently with cells in there somehow? | 19:26 |
| mriedem | that's what i'm wondering | 19:26 |
| alaski | cells isn't doing anything at that level | 19:26 |
| dansmith | okay | 19:26 |
| dansmith | hrm | 19:26 |
| dansmith | well, I guess we'll see | 19:27 |
| dansmith | I can't think of anything else that could cause that about this change | 19:27 |
| alaski | I can't either | 19:27 |
| dansmith | does anyone know where the open-lemming commit is? | 19:30 |
| dansmith | I can't find it | 19:30 |
| mriedem | yeah sec | 19:30 |
| *** iamjarvo has joined #openstack-nova | 19:30 | |
| *** iamjarvo has quit IRC | 19:31 | |
| mriedem | wtf, can't find it either | 19:31 |
| mriedem | i thought it was john's? | 19:31 |
| *** iamjarvo has joined #openstack-nova | 19:31 | |
| *** iamjarvo has quit IRC | 19:32 | |
| dansmith | I tried looking for him | 19:32 |
| dansmith | er, his | 19:32 |
| mriedem | https://review.openstack.org/#/q/owner:john%2540johngarbutt.com+status:open,n,z | 19:32 |
| dansmith | ttx: ? | 19:32 |
| alaski | https://review.openstack.org/#/c/171078/ | 19:32 |
| dansmith | ah, thanks | 19:32 |
| *** iamjarvo has joined #openstack-nova | 19:32 | |
| mriedem | alaski: dansmith: this doesn't have anything to do with that args weirdness? http://git.openstack.org/cgit/openstack/nova/tree/nova/compute/cells_api.py#n411 | 19:33 |
| dansmith | hmm | 19:34 |
| alaski | mriedem: if anything I would expect that to cause problems in the nova-cells service | 19:34 |
| *** burt has joined #openstack-nova | 19:35 | |
| alaski | I'm actually confused on how this got to the compute service at this point | 19:36 |
| openstackgerrit | Dan Smith proposed openstack/nova: Reserve 10 migrations for backports https://review.openstack.org/172185 | 19:38 |
| dansmith | that ^ needs to be the first commit in lemming, so if we could have it pre-approved, that'd be nice, kthxbai | 19:39 |
| *** Marga_ has quit IRC | 19:39 | |
| *** Marga_ has joined #openstack-nova | 19:40 | |
| *** Marga_ has quit IRC | 19:40 | |
| *** Marga_ has joined #openstack-nova | 19:40 | |
| dansmith | alaski: the weird thing is that that is making a call without kwargs, and with a different parameter name | 19:42 |
| dansmith | volume vs. volume_id | 19:42 |
| *** ijw has quit IRC | 19:43 | |
| mriedem | dansmith: that's actually 11 migrations, kthx | 19:44 |
| mriedem | :P | 19:44 |
| dansmith | dammit, I did that last time | 19:44 |
| mriedem | i remember | 19:44 |
| mriedem | there was much confusion in the irc's | 19:44 |
| alaski | so http://git.openstack.org/cgit/openstack/nova/tree/nova/compute/cells_api.py#n411 will end up calling into compute/api.py in the nova-cells service in the child, and from there there's nothing cells specific | 19:44 |
| *** Marga_ has quit IRC | 19:45 | |
| mriedem | also http://git.openstack.org/cgit/openstack/nova/tree/nova/tests/unit/db/test_migrations.py#n177 | 19:45 |
| *** mmedvede has joined #openstack-nova | 19:45 | |
| dansmith | mriedem: ah, right, thanks | 19:46 |
| dansmith | so much work to do today and so little brain power | 19:46 |
| mriedem | just like, do some stretches and eat some ginseng | 19:46 |
| *** xyang1 has joined #openstack-nova | 19:46 | |
| *** johnchalekson has quit IRC | 19:47 | |
| openstackgerrit | Dan Smith proposed openstack/nova: Reserve 10 migrations for backports https://review.openstack.org/172185 | 19:48 |
| dansmith | mriedem: yeah, that'll fix me | 19:48 |
| melwitt | dansmith, alaski: so before the proxy, the call wasn't going through run_compute_api_method in cells/messaging.py (it seems) otherwise this would have been happening before too | 19:48 |
| alaski | melwitt: yeah, that's what's confusing me | 19:49 |
| *** dims__ is now known as dims | 19:50 | |
| alaski | actually, what really gets me is that if it's a dict it should have failed at http://git.openstack.org/cgit/openstack/nova/tree/nova/compute/api.py#n3060, not on the compute | 19:50 |
| *** Sukhdev has quit IRC | 19:52 | |
| dansmith | alaski: remember the args might be switched | 19:52 |
| dansmith | on the manager side | 19:52 |
| dansmith | although volume_id shouldn't be a dict | 19:53 |
| alaski | yeah | 19:53 |
| melwitt | alaski: isn't it cells that makes it a dict after that in run_compute_api_method? it pulls from the db then passes that along without making it into an object because it's not in EXPECTS_OBJECTS | 19:53 |
| alaski | melwitt: yeah. which is why I would expect it to fail earlier :) | 19:53 |
| melwitt | I'm having trouble understanding the ordering this is happening in :\ | 19:54 |
| melwitt | I was thinking first is compute/api.py in the api cell, then the cells overrides, then cells messaging calls compute/api.py in the cell | 19:55 |
| *** zzzeek has joined #openstack-nova | 19:55 | |
| alaski | compute/api.py partially runs in the api, and compute/cells_api takes over at _detach_volumes. then cells rpc->cells manager->cells messaging which calls compute/api.py | 19:56 |
| melwitt | nvm, I see what you're saying. hm | 19:56 |
| *** jecarey_ has joined #openstack-nova | 19:56 | |
| *** ociuhandu has quit IRC | 19:57 | |
| *** otter768 has joined #openstack-nova | 19:57 | |
| *** jecarey__ has joined #openstack-nova | 19:57 | |
| *** jecarey has quit IRC | 19:58 | |
| *** jecarey_ has quit IRC | 20:01 | |
| *** otter768 has quit IRC | 20:01 | |
| alaski | oh | 20:02 |
| alaski | @object_compat isn't working because instance is no longer a kwarg | 20:02 |
| bauzas | alaski: any point on the cells job ? | 20:02 |
| alaski | dansmith: ^ | 20:03 |
| bauzas | oh nvm, not related | 20:03 |
| *** annashen has joined #openstack-nova | 20:03 | |
| dansmith | alaski: damn | 20:03 |
| *** ozialien has quit IRC | 20:03 | |
| alaski | there's still a cells fix needed, https://review.openstack.org/#/c/79741/, but that's the regression cause I think | 20:04 |
| dansmith | alaski: any easy way to determine if there are others? | 20:04 |
| dansmith | I guess we get this in and watch the cells job | 20:04 |
| alaski | yeah, I'll look at the tempest job | 20:04 |
| *** jichen has quit IRC | 20:05 | |
| dansmith | alaski: there | 20:05 |
| openstackgerrit | Dan Smith proposed openstack/nova: Add compute RPC API v4.0 https://review.openstack.org/172152 | 20:05 |
| melwitt | alaski: I notice in compute/rpcapi.py detach_volume calls with the volume_id, and that signature exists in compute/manager.py but not in compute/api.py (where it's just volume) | 20:05 |
| *** tonytan4ever has quit IRC | 20:07 | |
| alaski | melwitt: yeah, but the cells override handles that http://git.openstack.org/cgit/openstack/nova/tree/nova/compute/cells_api.py#n411 | 20:07 |
| *** lpetrut has quit IRC | 20:08 | |
| bauzas | alaski: I'm lurking the convo, and I see that you're looking at the Tempest job, right ? | 20:10 |
| alaski | bauzas: I was, but the network errors keep that from being too fruitful | 20:11 |
| bauzas | alaski: yeah, I think that unfortunately, we need to first merge melwitt's patches | 20:11 |
| mriedem | bauzas: melwitt's changes are ready to go | 20:11 |
| bauzas | cool | 20:12 |
| mriedem | oops devstack change failed | 20:12 |
| mriedem | https://review.openstack.org/#/c/171865/ | 20:12 |
| alaski | those are different issues | 20:13 |
| alaski | the target keeps moving on us | 20:13 |
| mriedem | keyerror | 20:13 |
| mriedem | http://logs.openstack.org/65/171865/4/experimental/check-tempest-dsvm-cells/7f21346/logs/screen-n-cell-child.txt.gz?level=TRACE | 20:13 |
| mriedem | yeah | 20:13 |
| alaski | fixes 43 failures though | 20:14 |
| *** tonytan4ever has joined #openstack-nova | 20:14 | |
| *** signed8bit has joined #openstack-nova | 20:14 | |
| mriedem | time to logstash fu | 20:14 |
| *** ijw has joined #openstack-nova | 20:14 | |
| *** iamjarvo has quit IRC | 20:15 | |
| mriedem | must not be indexed? | 20:17 |
| *** ivasev has quit IRC | 20:18 | |
| *** moshele has joined #openstack-nova | 20:19 | |
| dansmith | mriedem: I must've messed something else up | 20:19 |
| dansmith | I'm failing all the greandes | 20:19 |
| dansmith | or grenades even | 20:20 |
| melwitt | mriedem: I think not. :( I didn't know what's the criteria for getting indexed, that is I assumed it's not for a reason | 20:20 |
| mriedem | melwitt: well i know the cells job is indexed for nova at least | 20:21 |
| mriedem | might not be indexed for devstack since it's the experimental queue | 20:21 |
| *** ijw has quit IRC | 20:21 | |
| melwitt | oh, I guess I'm surprised this isn't already in the nova ones then | 20:22 |
| *** mwagner_lap has quit IRC | 20:22 | |
| mriedem | yeah.... | 20:23 |
| *** mspreitz has quit IRC | 20:23 | |
| *** achanda has quit IRC | 20:24 | |
| mriedem | not hitting message:"Error processing message locally" anywhere else though | 20:26 |
| mriedem | melwitt: could it be a thing with your change? | 20:26 |
| *** ozialien has joined #openstack-nova | 20:26 | |
| melwitt | mriedem: it could be if it's the same test that fails further down the line after the network part succeeds | 20:27 |
| mriedem | also seeing this in the logs http://goo.gl/nXCAEe | 20:30 |
| mriedem | which comes from nova.compute.api._local_delete | 20:30 |
| mriedem | it's not just in the cells job though | 20:30 |
| mriedem | melwitt: i'm guessing it's just a result of http://logs.openstack.org/65/171865/4/experimental/check-tempest-dsvm-cells/7f21346/logs/screen-n-cpu.txt.gz?level=TRACE#_2015-04-09_19_22_08_848 | 20:33 |
| mriedem | so shit blew up for awhile | 20:33 |
| melwitt | heh | 20:34 |
| mriedem | i'll recheck | 20:34 |
| bauzas | so I'm looking at http://logs.openstack.org/65/171865/4/experimental/check-tempest-dsvm-cells/7f21346/console.html and seeing only 4 errors left | 20:34 |
| *** ozialien has quit IRC | 20:34 | |
| mriedem | bauzas: that's what we're talking about | 20:35 |
| mriedem | i think those might be unrelated | 20:35 |
| bauzas | cool, sorry for being jetlagged | 20:35 |
| *** vladikr has quit IRC | 20:35 | |
| mriedem | http://goo.gl/FIJzyZ | 20:35 |
| mriedem | the service went away | 20:35 |
| mriedem | tests failed | 20:35 |
| mriedem | it came back up | 20:35 |
| mriedem | things started passing | 20:35 |
| bauzas | I count 2 of those related to boto | 20:35 |
| bauzas | mriedem: yeah I think it's unrelated | 20:36 |
| mriedem | it's not a block of tests running at the same time though from what i see | 20:36 |
| *** Marga_ has joined #openstack-nova | 20:36 | |
| mriedem | usually when we see model server went away, it's in grenade jobs where it's expected | 20:36 |
| *** baoli has joined #openstack-nova | 20:36 | |
| bauzas | have you checked req-id and timestamps ? | 20:37 |
| *** Marga_ has quit IRC | 20:38 | |
| *** igordcard_ has joined #openstack-nova | 20:38 | |
| bauzas | ok, digging into test_reset_state_server | 20:38 |
| *** hdd has joined #openstack-nova | 20:38 | |
| dansmith | hmm, is the grenade test broken? | 20:39 |
| dansmith | two of my patches are now failing it | 20:39 |
| dansmith | unrelated patches | 20:39 |
| mriedem | dansmith: log link? | 20:39 |
| dansmith | waiting for it | 20:39 |
| dansmith | but the other nova patch in the gate and check are failing them too it looks like | 20:39 |
| dansmith | both grenade jobs | 20:39 |
| dansmith | sdague: ^ | 20:39 |
| mriedem | grenade was refactored this week right? | 20:40 |
| dansmith | this one for example: https://review.openstack.org/#/c/171911/ | 20:40 |
| dansmith | mriedem: right | 20:40 |
| dansmith | that ^ one is failing in check right now, not reported yet | 20:40 |
| dansmith | one just passed jenkins about 20 minutes ago | 20:41 |
| mriedem | unrelated, but we should probably have infra make the xen ci job non-voting | 20:42 |
| mriedem | it's been over 24 hours of straight fail | 20:42 |
| mriedem | i guess i can wait 18 minutes to rant about that in the nova meeting | 20:42 |
| mikal | . | 20:42 |
| mriedem | you! | 20:43 |
| *** iamjarvo has joined #openstack-nova | 20:43 | |
| mikal | Hi? | 20:43 |
| mriedem | welcome to the jungle | 20:43 |
| dansmith | mriedem: https://jenkins02.openstack.org/job/check-grenade-dsvm/31687/console | 20:43 |
| dansmith | mriedem: failed in early stack.sh I think | 20:43 |
| mriedem | need grenade.sh | 20:43 |
| mikal | dansmith: do you have a review out for the 4.0 bump? | 20:43 |
| mriedem | he does | 20:43 |
| dansmith | mikal: yes | 20:43 |
| mriedem | https://review.openstack.org/#/c/172152/ | 20:43 |
| dansmith | mikal: I also have the migration placeholders u | 20:43 |
| dansmith | up | 20:44 |
| * mikal goes to take a look | 20:44 | |
| dansmith | mikal: we get to dump 40 compute RPC revisions if we do this now | 20:44 |
| mikal | It looks like once we've got those we could release | 20:44 |
| dansmith | yeah | 20:44 |
| dansmith | the migration ones comes post-release | 20:44 |
| dansmith | s/ones/one | 20:44 |
| *** ijw has joined #openstack-nova | 20:44 | |
| dansmith | so just the RPC one I think | 20:44 |
| *** adalbas has quit IRC | 20:44 | |
| mikal | Oh yeah, that | 20:45 |
| * mikal is just starting his first coffee | 20:45 | |
| dansmith | but I'm seeing a lot of grenade fail right now | 20:45 |
| dansmith | and that patch passed grenade before a recent revision | 20:45 |
| dansmith | so I'm concerned that we're going into a problem | 20:45 |
| mriedem | jogo: you around? | 20:45 |
| *** thangp has quit IRC | 20:46 | |
| *** andreykurilin_ has joined #openstack-nova | 20:46 | |
| mriedem | heh, all of our third party CI stats widgets don't work | 20:49 |
| *** baoli has quit IRC | 20:49 | |
| *** baoli has joined #openstack-nova | 20:49 | |
| *** smccully has quit IRC | 20:49 | |
| *** ijw has quit IRC | 20:51 | |
| jogo | mriedem: ? | 20:51 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Cells: Call compute api methods with instance objects https://review.openstack.org/79741 | 20:51 |
| mriedem | jogo: looks like grenade might be shitting the bed | 20:51 |
| dansmith | jogo: grenade seems broken | 20:51 |
| mriedem | jogo: trying to get some new results in check queue so we can logstash it | 20:51 |
| *** Marga_ has joined #openstack-nova | 20:51 | |
| jogo | http://jogo.github.io/gate/ | 20:51 |
| jogo | ahh partial-ncpu is going haywire | 20:52 |
| *** neelashah1 has quit IRC | 20:52 | |
| *** tpatil has quit IRC | 20:52 | |
| *** achanda has joined #openstack-nova | 20:53 | |
| jogo | mriedem: for third party CI https://github.com/jogo/lastcomment works pretty well | 20:53 |
| *** tpatil has joined #openstack-nova | 20:53 | |
| mriedem | CLI?! | 20:53 |
| *** iamjarvo has quit IRC | 20:53 | |
| *** sdake has quit IRC | 20:53 | |
| mriedem | http://goo.gl/RiRroc | 20:54 |
| mriedem | partial n-cpu last 24 hours fails | 20:54 |
| jogo | mriedem: CLI FTW | 20:54 |
| mriedem | http://logs.openstack.org/12/171912/3/gate/gate-grenade-dsvm-partial-ncpu/a8ba07e/logs/grenade.sh.txt.gz | 20:54 |
| mriedem | http://logs.openstack.org/12/171912/3/gate/gate-grenade-dsvm-partial-ncpu/a8ba07e/logs/grenade.sh.txt.gz#_2015-04-09_20_12_35_132 | 20:54 |
| dansmith | jogo: so, this is going to prevent us from landing our RPC bump :( | 20:54 |
| jogo | mriedem: build_name:"gate-grenade-dsvm-partial-ncpu" AND message:"Finished: FAILURE" | 20:54 |
| mriedem | did someone do an oslo release? | 20:55 |
| mriedem | http://logs.openstack.org/12/171912/3/gate/gate-grenade-dsvm-partial-ncpu/a8ba07e/logs/grenade.sh.txt.gz#_2015-04-09_20_12_32_007 | 20:55 |
| jogo | sdague: ^ grenade is having issues | 20:56 |
| jogo | dims: ^ | 20:56 |
| bauzas | alaski: around for a question ? | 20:57 |
| bauzas | alaski: http://logs.openstack.org/65/171865/4/experimental/check-tempest-dsvm-cells/7f21346/logs/screen-n-api.txt.gz#_2015-04-09_19_10_11_145 | 20:57 |
| bauzas | alaski: related to http://logs.openstack.org/65/171865/4/experimental/check-tempest-dsvm-cells/7f21346/console.html#_2015-04-09_19_30_11_552 | 20:58 |
| mriedem | here we go | 20:58 |
| mriedem | https://pypi.python.org/pypi/python-neutronclient | 20:58 |
| mriedem | jogo: dansmith: ^ | 20:58 |
| * jogo runs sdague's what broke script | 20:58 | |
| mriedem | https://pypi.python.org/pypi/python-neutronclient/2.4.0 | 20:58 |
| mriedem | neutronclient | 20:58 |
| mriedem | rleeased today | 20:58 |
| mriedem | mestery: ^ | 20:58 |
| *** iamjarvo has joined #openstack-nova | 20:58 | |
| bauzas | alaski: why are we calling the compute api instead of the cells api ? | 20:58 |
| dansmith | mriedem: gah | 20:58 |
| mriedem | wtf | 20:58 |
| mriedem | ok, reporting bug | 20:58 |
| bauzas | alaski: sounds like it checks the main DB and not the cell DB | 20:58 |
| * jogo calls dibs on responding to the ML thread and calling em out | 20:59 | |
| mestery | mriedem: WE have a patch for that: https://review.openstack.org/#/c/172150/ | 20:59 |
| mestery | mriedem: amotoki replied to the version capping thread with the plan we had, if we've messed stuff up, I apologize in advance. | 20:59 |
| dansmith | jogo: do eet | 20:59 |
| dims | jogo: https://pypi.python.org/pypi/python-neutronclient/2.4.0 ? | 21:00 |
| jogo | dims: yeah, I pinged you since thought it may have been oslo related | 21:00 |
| mikal | Meeting? | 21:00 |
| jogo | mriedem: you have a bug number? | 21:00 |
| mriedem | yes | 21:00 |
| mikal | Or shall we all just grumble in our corners? | 21:00 |
| melwitt | let's meet! | 21:00 |
| mriedem | jogo: no bug yet | 21:00 |
| mriedem | we need one | 21:00 |
| mriedem | sec | 21:00 |
| mriedem | https://bugs.launchpad.net/python-neutronclient/+bug/1442367 | 21:02 |
| openstack | Launchpad bug 1442367 in python-neutronclient "2.4.0 release destroys grenade since stable/juno isn't capped properly" [Critical,Confirmed] | 21:02 |
| alaski | bauzas: the cells api subclasses the compute api but doesn't implement that method, so it shows up as the compute api | 21:02 |
| bauzas | alaski: so that's probably why it fails | 21:03 |
| alaski | bauzas: the get should be fine, it's just pulling the instance from the db. nothing cells specific needed | 21:03 |
| bauzas | alaski: lemme understand correctly | 21:04 |
| bauzas | alaski: instances are not cell-related ? | 21:04 |
| *** mwagner_lap has joined #openstack-nova | 21:04 | |
| jogo | mriedem: grrr you beat me | 21:04 |
| *** hdd has quit IRC | 21:05 | |
| *** emagana has joined #openstack-nova | 21:06 | |
| *** rfolco has quit IRC | 21:07 | |
| alaski | bauzas: I mean pulling an instance from the db for the api doesn't need to involve cells | 21:08 |
| *** aysyd has quit IRC | 21:08 | |
| alaski | bauzas: cells stores the instance in both dbs so that it can be pulled like that | 21:08 |
| bauzas | alaski: and saving ? | 21:09 |
| alaski | bauzas: saving replicates the update down/up | 21:10 |
| bauzas | alaski: https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/contrib/admin_actions.py#L378-L382 | 21:10 |
| *** pkoniszewski has quit IRC | 21:11 | |
| *** jcoufal has joined #openstack-nova | 21:12 | |
| *** raildo has quit IRC | 21:12 | |
| alaski | bauzas: https://github.com/openstack/nova/blob/master/nova/compute/__init__.py#L37 that should be the right compute_api | 21:12 |
| alaski | bauzas: and then the 'get' runs in compute/api because of https://github.com/openstack/nova/blob/master/nova/compute/cells_api.py#L162 | 21:13 |
| bauzas | alaski: but it gets the instance get method from the common module | 21:13 |
| alaski | bauzas: https://github.com/openstack/nova/blob/master/nova/api/openstack/common.py#L531 | 21:14 |
| alaski | bauzas: are you saying it's getting the wrong instance somehow? | 21:14 |
| bauzas | alaski: that's what I'm wondering | 21:15 |
| alaski | I'm not sure how it could. and we get a lot of instances that way | 21:15 |
| bauzas | alaski: well, I'm just wondering how it can fail, because that's just a pure API <-> DB | 21:16 |
| bauzas | alaski: like update my state and bye bye | 21:16 |
| bauzas | alaski: but for some reason, it doesn't update it | 21:17 |
| alaski | bauzas: gotcha, let me look at the other logs again | 21:18 |
| *** rushil has quit IRC | 21:24 | |
| *** dims has quit IRC | 21:24 | |
| *** rushil has joined #openstack-nova | 21:25 | |
| dansmith | mriedem: we're waiting for 172150 to merge before we can land anything right? | 21:30 |
| mriedem | dansmith: well, that and all of the syncs | 21:30 |
| mriedem | so it's going to be awhile | 21:30 |
| dansmith | mriedem: is it in gate now? | 21:30 |
| mriedem | yes | 21:31 |
| mriedem | nearly done | 21:31 |
| dansmith | okay | 21:31 |
| *** r-daneel has joined #openstack-nova | 21:31 | |
| mikal | jogo: its pleasing to me how much of a psychological impact the phrase "early mark" has had on you | 21:34 |
| *** rushiagr_away has quit IRC | 21:35 | |
| jogo | haha | 21:35 |
| * mikal wanders off to get out of PJs | 21:35 | |
| mriedem | and into a dry martini | 21:36 |
| *** iamjarvo has quit IRC | 21:37 | |
| tjones1 | :-D | 21:38 |
| *** tpatil has quit IRC | 21:40 | |
| *** gyee has joined #openstack-nova | 21:41 | |
| *** Sukhdev has joined #openstack-nova | 21:45 | |
| *** ijw has joined #openstack-nova | 21:47 | |
| bauzas | alaski: still diggin | 21:48 |
| alaski | bauzas: me too. I can see that the state gets set correctly, but there are messages coming up from the cell to set the state to error | 21:48 |
| alaski | bauzas: they're part of the build request, but that succeeds | 21:49 |
| bauzas | alaski: so reset_state calls common.get_instance() which itself calls compute_api.get() | 21:49 |
| bauzas | but IIUC, compute_api is actually hooked on cells_api right? | 21:49 |
| bauzas | so, why can't I find a get() method on cells_api ? | 21:49 |
| alaski | cells_api subclasses compute_api, but doesn't override everything | 21:50 |
| bauzas | alaski: I understand it, so I means we get the instance from the main DB | 21:51 |
| bauzas | like you said | 21:51 |
| *** andreykurilin_ has quit IRC | 21:51 | |
| *** ijw has quit IRC | 21:52 | |
| dansmith | mriedem: that merged | 21:53 |
| dansmith | mriedem: what should I track now? | 21:53 |
| *** rushiagr_away has joined #openstack-nova | 21:54 | |
| bauzas | alaski: ok, seeing the cell conditional on the instance.save() method | 21:55 |
| mriedem | dansmith: there should be a global reqs update to stable | 21:55 |
| *** isd has quit IRC | 21:55 | |
| alaski | bauzas: http://logs.openstack.org/65/171865/4/experimental/check-tempest-dsvm-cells/7f21346/logs/screen-n-cell-region.txt.gz#_2015-04-09_19_10_11_392 is an update to error, and right after that req-cf3cd691-e2f2-49af-9469-6f4b424da24e sets it to active | 21:56 |
| alaski | bauzas: but the GET comes in between those two | 21:56 |
| alaski | bauzas: the question is what triggered the error update, because it' wasn't the earlier reset-state | 21:56 |
| dansmith | mriedem: this? https://review.openstack.org/#/c/172219/ | 21:57 |
| dansmith | mriedem: adam_g was on that one so hard | 21:57 |
| mriedem | yeah | 21:57 |
| cfriesen | So I just realized I missed the nova meeting. Is it absolutely too late for kilo? I think https://review.openstack.org/170780 (adding in qemu version checks for NUMA and hugepages) would make sense since that stuff is pretty new. We recently added libvirt version checks, but not qemu checks. | 21:57 |
| dansmith | cfriesen: yes | 21:57 |
| dansmith | cfriesen: nothing other than the RPC bump and then the tag | 21:57 |
| mriedem | cfriesen: there are always stable backports | 21:58 |
| *** otter768 has joined #openstack-nova | 21:58 | |
| cfriesen | mriedem: true enough | 21:58 |
| mriedem | or kilo-proposed cherry pick if it's important enough | 21:58 |
| mriedem | for an rc2 | 21:58 |
| dansmith | yeah, but I think a backport is better for this one | 21:58 |
| *** sbfox has quit IRC | 21:58 | |
| *** moshele has quit IRC | 22:00 | |
| alaski | bauzas: oh wait, it is the reset-state from before | 22:00 |
| alaski | bauzas: I think the first reset-state updates the api db, then the child, then propogates back up. meanwhile the second reset-state updates the api db, then gets trampled when the first one propogates up | 22:01 |
| bauzas | alaski: yeah | 22:02 |
| *** otter768 has quit IRC | 22:03 | |
| *** emagana has quit IRC | 22:03 | |
| *** iamjarvo has joined #openstack-nova | 22:03 | |
| *** burt has quit IRC | 22:04 | |
| mikal | . | 22:04 |
| *** sbfox has joined #openstack-nova | 22:06 | |
| bauzas | alaski: I need to jump straight on how it's propagated | 22:06 |
| *** dboik_ has joined #openstack-nova | 22:06 | |
| *** sdake has joined #openstack-nova | 22:07 | |
| openstackgerrit | Artom Lifshitz proposed openstack/nova-specs: Support --all-tenants in floatingips-list https://review.openstack.org/171389 | 22:07 |
| *** artom has quit IRC | 22:08 | |
| alaski | bauzas: follow the object.save | 22:09 |
| *** dboik has quit IRC | 22:09 | |
| *** dboik_ has quit IRC | 22:10 | |
| *** sdake_ has joined #openstack-nova | 22:10 | |
| *** Guest77301 has joined #openstack-nova | 22:11 | |
| bauzas | alaski: yup, that's what I read... update_at_top | 22:11 |
| bauzas | so, racy ? | 22:12 |
| alaski | bauzas: yeah :| | 22:12 |
| *** Guest77301 is now known as dims_ | 22:12 | |
| alaski | bauzas: or eventually consistent :) | 22:13 |
| dansmith | lol | 22:13 |
| dansmith | they are the same thing, right? depending on your slant :) | 22:13 |
| alaski | exactly :) | 22:13 |
| bauzas | lol | 22:13 |
| *** sdake has quit IRC | 22:14 | |
| bauzas | eventually means probably in French, that sounds really like it :) | 22:14 |
| *** jcoufal has quit IRC | 22:14 | |
| bauzas | that's a common mistake for english begineers | 22:14 |
| dansmith | in english it means it will happen, probably is less strong | 22:14 |
| *** eharney has quit IRC | 22:15 | |
| bauzas | dansmith: yeah, was just kidding on the 'probably consistent'' which would mean "try it, if you're lucky" | 22:15 |
| *** VW_ has quit IRC | 22:15 | |
| dansmith | heh | 22:15 |
| bauzas | sooooo | 22:16 |
| bauzas | I don't like putting a semaphore on top of this | 22:17 |
| *** Nic has quit IRC | 22:18 | |
| bauzas | the problem is that we're updating down before updating up | 22:18 |
| *** annashen has quit IRC | 22:18 | |
| *** adrian_otto has joined #openstack-nova | 22:19 | |
| alaski | I think the real question is why are we updating back up | 22:21 |
| bauzas | alaski: agreed | 22:21 |
| *** Nic has joined #openstack-nova | 22:21 | |
| *** dave-mccowan has quit IRC | 22:22 | |
| melwitt | it would have to be something in the cell calling instance.save(), that would cause it to go back up | 22:22 |
| bauzas | melwitt: yeah but we should note that the cell save is already coming from north | 22:23 |
| alaski | melwitt: yeah. sorry, I meant more why don't we stop that | 22:23 |
| bauzas | melwitt: that back-and-worth is bad, we should only update straight one direction | 22:23 |
| alaski | melwitt: in the same way we want to stop it from looping when we save with an object in the api from a cell update | 22:23 |
| bauzas | either northbound or southbound | 22:24 |
| melwitt | okay, I was thinking maybe it's not back and forth. that some thing happens inside the cell such that it wants to save the instance (like some other state transition) and then it will go back up | 22:24 |
| mriedem | dansmith: think i found something here https://review.openstack.org/#/c/172152/ | 22:25 |
| *** oro has quit IRC | 22:25 | |
| mriedem | in reserve_block_device_name | 22:25 |
| bauzas | melwitt: sure, if the compute is calling instance.save(), we need to update the api db | 22:25 |
| melwitt | bauzas alaski: unless you're saying that you have already determined compute isn't calling save() | 22:26 |
| bauzas | melwitt: but if the api is calling instance.save(), we should prevent this back-and-forth when going down to the compute | 22:26 |
| bauzas | s/compute/cell | 22:26 |
| *** annashen has joined #openstack-nova | 22:26 | |
| alaski | melwitt: in this case it's not compute, it's instance_update_from_api | 22:26 |
| melwitt | bauzas alaski: oh :( | 22:27 |
| alaski | melwitt: yeah. so we still need a solution like we were thinking about for adding instance.save to instance_update_at_top | 22:27 |
| bauzas | just to be pragmatic, while we're thinking at a possible solution, could we consider whitelisting this one ? | 22:28 |
| melwitt | alaski: gotcha. | 22:29 |
| *** FL1SK has quit IRC | 22:29 | |
| bauzas | ok, time is up | 22:29 |
| dansmith | mriedem: I think it's right. see if my explanation makes sense | 22:29 |
| bauzas | see you all tomorrow | 22:30 |
| *** tjones1 has quit IRC | 22:30 | |
| alaski | bauzas: that would be good, if we open a bug to link to so we know to bring the test back when this is fixed | 22:30 |
| alaski | bauzas: bye | 22:30 |
| melwitt | seeya bauzas | 22:30 |
| bauzas | alaski: I would be tempted to whitelist those 4, and mark those as bugs so we keep track of them | 22:30 |
| *** vilobhmm1 has left #openstack-nova | 22:31 | |
| bauzas | alaski: so we can continue to work on them while the job would be green - and voting | 22:31 |
| alaski | bauzas: that would be nice | 22:31 |
| melwitt | my worry is if we have a race, other failures might pop up that aren't these 4 | 22:31 |
| melwitt | but that's just a side note | 22:31 |
| bauzas | alaski: I'm concerned by any tunnel effect which could consistently push back our efforts to make the job votion | 22:31 |
| bauzas | voting | 22:31 |
| bauzas | melwitt: that's a race because the test is done that way | 22:31 |
| bauzas | melwitt: that's a flip | 22:32 |
| melwitt | bauzas: okay | 22:32 |
| melwitt | if it's related to how those particular tests work, that's good. makes it more unlikely we'd see it in others | 22:32 |
| bauzas | melwitt: so I don't think that's something really terrible, just something that can be raising up that we need to fix soon | 22:32 |
| bauzas | melwitt: the test is marking the instance state to ERROR and then back ACTIVE | 22:33 |
| melwitt | bauzas: I see | 22:33 |
| *** emagana has joined #openstack-nova | 22:33 | |
| alaski | well, as much as I'd like the job voting. I would like to see it green for a bit first, to shake out any races we're going to hit | 22:33 |
| bauzas | melwitt: the race is when the ACTIVE state is wiped by the ERROR return | 22:33 |
| bauzas | alaski: sure, my point is that we need to keep it stable | 22:34 |
| bauzas | and having it red doesn't help that | 22:34 |
| mriedem | dansmith: alright, i think i'm +W on this then | 22:34 |
| *** jecarey__ has quit IRC | 22:34 | |
| alaski | bauzas: agreed. let's get it green | 22:34 |
| melwitt | yeah, my thinking was more broad, if we know there's a race, it could be a coincidence that these tests are the failed ones this time, next time it could be others. but if certain tests make it more likely, then I understand that | 22:35 |
| *** emagana_ has joined #openstack-nova | 22:35 | |
| *** rushil has quit IRC | 22:35 | |
| bauzas | melwitt: agreed, that's not a perfect situation, but at least we know the problem | 22:35 |
| melwitt | yeah | 22:35 |
| bauzas | melwitt: if other tests are failing, we can see if they reproduce the behaviour of updating an instance and then updating it again | 22:36 |
| *** signed8bit has quit IRC | 22:36 | |
| dansmith | mriedem: cool, thanks | 22:38 |
| mriedem | dansmith: do we care about any of the other jobs? we know grenade is borked | 22:38 |
| *** emagana has quit IRC | 22:38 | |
| mriedem | i wasn't sure about cells | 22:38 |
| *** emagana_ has quit IRC | 22:39 | |
| mriedem | seems that logs/screen-n-cell-child.txt isn't indexed so i can't tell when this showed up | 22:40 |
| alaski | mriedem: until melwitts changes merged there's a lot of noise in the cells job. but it's at 47 failures which is where it was before | 22:40 |
| alaski | s/merged/merge/ | 22:40 |
| mriedem | alaski: i'm kind of more concerned that we don't have the cells logs indexed.... | 22:40 |
| mriedem | let me see if i can find what's up with that | 22:40 |
| *** dave-mccowan has joined #openstack-nova | 22:41 | |
| mriedem | i suppose melwitt's change is dependent on the grenade failures too | 22:41 |
| mriedem | s/dependent/blocked/ | 22:41 |
| mriedem | plus +W on the devstack change | 22:41 |
| melwitt | mriedem: yes :( | 22:41 |
| mriedem | also, we shouldn't be running the cells job on stable probably | 22:42 |
| dansmith | mriedem: what do you mean? | 22:43 |
| dansmith | mriedem: on the rpc patch? | 22:43 |
| mriedem | dansmith: yeah http://logs.openstack.org/52/172152/4/check/check-tempest-dsvm-cells/b739bbd/logs/screen-n-cell-child.txt.gz?level=TRACE#_2015-04-09_20_47_46_048 | 22:43 |
| mriedem | but i saw that in melwitt's devstack change also | 22:43 |
| mriedem | so i guess it's just a new thing | 22:43 |
| dansmith | ah, hmm | 22:43 |
| *** Marga_ has quit IRC | 22:43 | |
| dansmith | is that the noise alaski is talking about? | 22:43 |
| mriedem | i think so | 22:44 |
| dansmith | okay | 22:44 |
| mriedem | w/o logstash i'm helpless :) | 22:44 |
| *** sbfox has quit IRC | 22:45 | |
| mriedem | yup, we don't index cells logs in system-config | 22:45 |
| *** emagana has joined #openstack-nova | 22:45 | |
| mriedem | alright, time to add that | 22:45 |
| * mriedem rolls up sleeves higher than they already are | 22:45 | |
| melwitt | that particular error isn't part of the noise. the noise is the network related failures, 43 of the 47 | 22:46 |
| *** baoli has quit IRC | 22:46 | |
| melwitt | I guess from your point of view it's noise, as it's expected to fail. it's not related to your change | 22:47 |
| *** sdake_ has quit IRC | 22:47 | |
| alaski | +1 | 22:47 |
| mriedem | ok, so i'm going to +W the rpc change then | 22:47 |
| alaski | alright, gotta run for dinner | 22:47 |
| mriedem | o/ | 22:47 |
| *** rushil has joined #openstack-nova | 22:47 | |
| melwitt | seeya alaski | 22:47 |
| *** ijw has joined #openstack-nova | 22:48 | |
| dansmith | mriedem: thanks | 22:49 |
| dansmith | I busted my ass to get that done this morning, glad it wasn't for nothing :) | 22:49 |
| *** rushil has quit IRC | 22:50 | |
| *** adrian_otto1 has joined #openstack-nova | 22:50 | |
| *** gyee has quit IRC | 22:51 | |
| *** adrian_otto has quit IRC | 22:52 | |
| *** dims_ has quit IRC | 22:52 | |
| mriedem | here we go https://review.openstack.org/#/c/172231/ | 22:52 |
| *** ijw has quit IRC | 22:52 | |
| mriedem | 172219 is in the gate | 22:53 |
| *** Nic has quit IRC | 22:54 | |
| *** Nic has joined #openstack-nova | 22:54 | |
| dansmith | aye | 22:57 |
| *** emagana has quit IRC | 22:58 | |
| *** hemna is now known as hemnafk | 23:02 | |
| *** claudiub has quit IRC | 23:02 | |
| *** zzzeek has quit IRC | 23:06 | |
| *** mriedem is now known as mriedem_away | 23:11 | |
| *** r-daneel has quit IRC | 23:13 | |
| *** Marga_ has joined #openstack-nova | 23:14 | |
| *** gyee has joined #openstack-nova | 23:16 | |
| *** ganso_ has quit IRC | 23:17 | |
| *** iamjarvo has quit IRC | 23:22 | |
| *** dboik has joined #openstack-nova | 23:29 | |
| *** dboik_ has joined #openstack-nova | 23:29 | |
| *** zzzeek has joined #openstack-nova | 23:32 | |
| *** annashen has quit IRC | 23:32 | |
| *** annashen has joined #openstack-nova | 23:32 | |
| *** dboik has quit IRC | 23:33 | |
| *** jcoufal has joined #openstack-nova | 23:34 | |
| *** annashen has quit IRC | 23:34 | |
| *** Marga_ has quit IRC | 23:35 | |
| *** Marga_ has joined #openstack-nova | 23:36 | |
| *** dims_ has joined #openstack-nova | 23:38 | |
| *** Marga_ has quit IRC | 23:38 | |
| *** IanGovett has quit IRC | 23:40 | |
| *** Sukhdev has quit IRC | 23:45 | |
| *** Marga_ has joined #openstack-nova | 23:47 | |
| *** Marga_ has quit IRC | 23:48 | |
| *** Marga_ has joined #openstack-nova | 23:48 | |
| *** ijw has joined #openstack-nova | 23:49 | |
| *** ijw has quit IRC | 23:53 | |
| dansmith | woot, rpc in the gate | 23:56 |
| *** otter768 has joined #openstack-nova | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!