| *** Sukhdev has quit IRC | 00:01 | |
| *** otter768 has joined #openstack-nova | 00:01 | |
| *** sdake has quit IRC | 00:03 | |
| *** apmelton_away is now known as apmelton | 00:05 | |
| *** tsekiyam_ has joined #openstack-nova | 00:06 | |
| *** salv-orlando has quit IRC | 00:07 | |
| *** achanda has quit IRC | 00:07 | |
| *** mtanino has quit IRC | 00:08 | |
| *** salv-orlando has joined #openstack-nova | 00:08 | |
| *** tsekiyama has quit IRC | 00:09 | |
| *** tsekiyam_ has quit IRC | 00:11 | |
| *** zzzeek has quit IRC | 00:11 | |
| *** Nikolay_St has quit IRC | 00:11 | |
| *** Nikolay_St has joined #openstack-nova | 00:12 | |
| *** apuimedo has quit IRC | 00:14 | |
| *** annegentle has joined #openstack-nova | 00:15 | |
| *** vilobhmm11 has quit IRC | 00:15 | |
| *** vilobhmm1 has joined #openstack-nova | 00:15 | |
| *** artom has joined #openstack-nova | 00:17 | |
| *** otter768 has quit IRC | 00:19 | |
| jroll | dansmith: pew pew, with diff: https://gist.github.com/jimrollenhagen/c6b4926d605d9b4e8da9#file-nova-compute-log-L132 | 00:19 |
|---|---|---|
| jroll | doesn't look like it changed, so I guess this is expected, must be something else clobbering it | 00:19 |
| *** annegentle has quit IRC | 00:20 | |
| *** thangp has joined #openstack-nova | 00:20 | |
| * jroll goes away for the night | 00:20 | |
| *** salv-orlando has quit IRC | 00:22 | |
| *** Marga_ has quit IRC | 00:23 | |
| *** patrickeast has quit IRC | 00:24 | |
| *** artom has quit IRC | 00:25 | |
| *** gyee has quit IRC | 00:29 | |
| *** Nikolay_St has quit IRC | 00:37 | |
| *** Nikolay_St has joined #openstack-nova | 00:38 | |
| *** ijw has joined #openstack-nova | 00:38 | |
| *** otter768 has joined #openstack-nova | 00:40 | |
| *** apuimedo has joined #openstack-nova | 00:42 | |
| *** isd1 has quit IRC | 00:43 | |
| *** Marga_ has joined #openstack-nova | 00:45 | |
| *** Marga_ has quit IRC | 00:46 | |
| *** nikhil_k is now known as nikhil_k-afk | 00:48 | |
| *** promulo_ has joined #openstack-nova | 00:48 | |
| *** promulo__ has quit IRC | 00:52 | |
| *** Marga_ has joined #openstack-nova | 00:53 | |
| *** Marga_ has quit IRC | 00:54 | |
| *** bkopilov has quit IRC | 00:58 | |
| *** jyuso1 has joined #openstack-nova | 01:03 | |
| *** ijw has quit IRC | 01:05 | |
| *** ijw has joined #openstack-nova | 01:06 | |
| *** Nikolay_St has quit IRC | 01:07 | |
| *** Nikolay_St has joined #openstack-nova | 01:07 | |
| *** apuimedo has quit IRC | 01:12 | |
| *** haigang has joined #openstack-nova | 01:15 | |
| *** ijw has quit IRC | 01:16 | |
| *** signed8bit has quit IRC | 01:17 | |
| *** popw has joined #openstack-nova | 01:17 | |
| *** joefides_ has quit IRC | 01:17 | |
| *** bkopilov has joined #openstack-nova | 01:18 | |
| *** joefides has joined #openstack-nova | 01:18 | |
| *** baoli has quit IRC | 01:18 | |
| *** sdake has joined #openstack-nova | 01:22 | |
| *** sdake_ has quit IRC | 01:26 | |
| *** flaper87 has quit IRC | 01:27 | |
| *** emagana has quit IRC | 01:27 | |
| *** flaper87 has joined #openstack-nova | 01:28 | |
| *** liusheng has joined #openstack-nova | 01:29 | |
| *** Nikolay_St has quit IRC | 01:32 | |
| *** Nikolay_St has joined #openstack-nova | 01:33 | |
| *** flaper87 has quit IRC | 01:33 | |
| *** sorrison has quit IRC | 01:33 | |
| *** salv-orlando has joined #openstack-nova | 01:35 | |
| *** greghaynes has joined #openstack-nova | 01:35 | |
| *** erkules_ has joined #openstack-nova | 01:36 | |
| *** nelsnelson has joined #openstack-nova | 01:38 | |
| *** erkules has quit IRC | 01:38 | |
| *** apuimedo has joined #openstack-nova | 01:39 | |
| *** flaper87 has joined #openstack-nova | 01:39 | |
| *** vilobhmm1 has quit IRC | 01:40 | |
| *** sorrison has joined #openstack-nova | 01:40 | |
| *** kaisers has quit IRC | 01:40 | |
| *** vilobhmm1 has joined #openstack-nova | 01:40 | |
| *** vilobhmm1 has quit IRC | 01:40 | |
| *** nelsnelson has quit IRC | 01:41 | |
| *** kaisers has joined #openstack-nova | 01:42 | |
| *** nelsnelson has joined #openstack-nova | 01:43 | |
| *** salv-orlando has quit IRC | 01:45 | |
| *** sdake_ has joined #openstack-nova | 01:50 | |
| *** sdake has quit IRC | 01:53 | |
| *** garyk has quit IRC | 01:54 | |
| *** garyk has joined #openstack-nova | 01:55 | |
| *** asettle is now known as asettle-luncheon | 01:57 | |
| *** apmelton is now known as apmelton_away | 01:57 | |
| openstackgerrit | Chris Friesen proposed openstack/nova: unify libvirt driver checks for qemu https://review.openstack.org/175542 | 01:57 |
| *** Nikolay_St has quit IRC | 01:58 | |
| *** Nikolay_St has joined #openstack-nova | 01:59 | |
| *** harlowja is now known as harlowja_away | 02:02 | |
| *** xianghuihui has joined #openstack-nova | 02:02 | |
| *** bkopilov has quit IRC | 02:03 | |
| *** baoli has joined #openstack-nova | 02:03 | |
| *** kbyrne has quit IRC | 02:05 | |
| *** mwagner_lap has quit IRC | 02:07 | |
| *** xianghuihui has quit IRC | 02:09 | |
| *** kbyrne has joined #openstack-nova | 02:10 | |
| *** baoli has quit IRC | 02:10 | |
| *** baoli has joined #openstack-nova | 02:12 | |
| *** zhangtralon has joined #openstack-nova | 02:14 | |
| *** rfolco has quit IRC | 02:15 | |
| *** READ10 has quit IRC | 02:15 | |
| *** salv-orlando has joined #openstack-nova | 02:15 | |
| *** baoli has quit IRC | 02:19 | |
| *** salv-orlando has quit IRC | 02:20 | |
| *** mwagner_lap has joined #openstack-nova | 02:21 | |
| *** xyang1 has quit IRC | 02:23 | |
| *** Nikolay_St has quit IRC | 02:25 | |
| *** Nikolay_St has joined #openstack-nova | 02:26 | |
| *** thangp has quit IRC | 02:33 | |
| *** jyuso1 has quit IRC | 02:37 | |
| *** marun has joined #openstack-nova | 02:37 | |
| *** achanda has joined #openstack-nova | 02:37 | |
| *** unicell has quit IRC | 02:39 | |
| *** baoli has joined #openstack-nova | 02:40 | |
| *** baoli_ has joined #openstack-nova | 02:42 | |
| *** baoli has quit IRC | 02:42 | |
| *** wuhg has quit IRC | 02:49 | |
| *** VW_ has joined #openstack-nova | 02:55 | |
| *** haigang has quit IRC | 02:58 | |
| *** sdake has joined #openstack-nova | 03:00 | |
| *** marun has quit IRC | 03:01 | |
| *** annashen has joined #openstack-nova | 03:02 | |
| *** Murali has joined #openstack-nova | 03:02 | |
| *** xylan has left #openstack-nova | 03:03 | |
| *** sdake_ has quit IRC | 03:04 | |
| *** sdake_ has joined #openstack-nova | 03:04 | |
| *** sdake has quit IRC | 03:08 | |
| *** annashen has quit IRC | 03:14 | |
| *** popw1 has joined #openstack-nova | 03:14 | |
| *** BadCub is now known as BadCub_Hiding | 03:15 | |
| *** popw has quit IRC | 03:16 | |
| *** dikonoor has joined #openstack-nova | 03:17 | |
| *** Nikolay_St has quit IRC | 03:20 | |
| *** Nikolay_St has joined #openstack-nova | 03:21 | |
| *** VW_ has quit IRC | 03:21 | |
| *** VW_ has joined #openstack-nova | 03:22 | |
| *** Murali has quit IRC | 03:22 | |
| *** yamahata has joined #openstack-nova | 03:23 | |
| *** Longgeek has joined #openstack-nova | 03:33 | |
| *** dave-mccowan has quit IRC | 03:36 | |
| *** rushiagr_away is now known as rushiagr | 03:40 | |
| openstackgerrit | Vincent Hou proposed openstack/nova: Correct the order of the parameters to call swap_volume https://review.openstack.org/174307 | 03:40 |
| *** xianghuihui has joined #openstack-nova | 03:40 | |
| *** xianghuihui has quit IRC | 03:40 | |
| *** xianghuihui has joined #openstack-nova | 03:41 | |
| *** xianghuihui has quit IRC | 03:41 | |
| *** achanda has quit IRC | 03:44 | |
| *** dikonoor has quit IRC | 03:44 | |
| *** Longgeek has quit IRC | 03:51 | |
| *** VW_ has quit IRC | 03:52 | |
| *** otter768 has quit IRC | 03:54 | |
| *** asettle-luncheon is now known as asettle | 03:54 | |
| *** VW_ has joined #openstack-nova | 03:56 | |
| *** alanf-mc has quit IRC | 03:56 | |
| *** tojuvone has joined #openstack-nova | 03:56 | |
| *** vilobhmm1 has joined #openstack-nova | 03:57 | |
| *** RuiChen has quit IRC | 03:58 | |
| *** RuiChen has joined #openstack-nova | 03:59 | |
| *** Longgeek has joined #openstack-nova | 04:00 | |
| *** rushiagr is now known as rushiagr_away | 04:00 | |
| *** rushiagr_away is now known as rushiagr | 04:02 | |
| *** baoli_ has quit IRC | 04:06 | |
| *** Nikolay_St has quit IRC | 04:07 | |
| *** Nikolay_St has joined #openstack-nova | 04:08 | |
| *** Nikolay_St has quit IRC | 04:14 | |
| *** Nikolay_St has joined #openstack-nova | 04:14 | |
| *** annashen has joined #openstack-nova | 04:15 | |
| *** rajesht has joined #openstack-nova | 04:17 | |
| *** salv-orlando has joined #openstack-nova | 04:18 | |
| *** annashen has quit IRC | 04:19 | |
| *** rajesht has quit IRC | 04:25 | |
| *** salv-orlando has quit IRC | 04:25 | |
| *** subscope_ has joined #openstack-nova | 04:30 | |
| *** Murali has joined #openstack-nova | 04:31 | |
| *** achanda has joined #openstack-nova | 04:35 | |
| *** kaisers has quit IRC | 04:37 | |
| *** Nikolay_St has quit IRC | 04:40 | |
| *** Nikolay_St has joined #openstack-nova | 04:41 | |
| *** zhangtralon has quit IRC | 04:41 | |
| *** Marga_ has joined #openstack-nova | 04:41 | |
| *** Longgeek has quit IRC | 04:42 | |
| *** unicell has joined #openstack-nova | 04:42 | |
| *** emagana has joined #openstack-nova | 04:45 | |
| *** takedakn has joined #openstack-nova | 04:45 | |
| *** takedakn has quit IRC | 04:45 | |
| *** markvoelker_ has quit IRC | 04:51 | |
| *** Longgeek has joined #openstack-nova | 04:52 | |
| *** wuhg has joined #openstack-nova | 04:52 | |
| *** nkrinner has joined #openstack-nova | 04:52 | |
| *** kaisers has joined #openstack-nova | 04:54 | |
| *** annashen has joined #openstack-nova | 04:54 | |
| *** Murali has quit IRC | 04:55 | |
| *** nelsnelson has quit IRC | 04:55 | |
| *** salv-orlando has joined #openstack-nova | 04:58 | |
| *** browne has joined #openstack-nova | 05:03 | |
| *** bkopilov has joined #openstack-nova | 05:04 | |
| *** salv-orlando has quit IRC | 05:05 | |
| *** Nikolay_St has quit IRC | 05:06 | |
| *** rushiagr is now known as rushiagr_away | 05:06 | |
| *** Nikolay_St has joined #openstack-nova | 05:07 | |
| *** Marga_ has quit IRC | 05:10 | |
| *** Marga_ has joined #openstack-nova | 05:11 | |
| *** kiran-r has joined #openstack-nova | 05:15 | |
| *** garyk has quit IRC | 05:16 | |
| *** salv-orlando has joined #openstack-nova | 05:17 | |
| *** sks has joined #openstack-nova | 05:18 | |
| *** salv-orlando has quit IRC | 05:21 | |
| *** redbeard has joined #openstack-nova | 05:22 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: WIP: Add ConsoleAuthToken object and console_auth_token table https://review.openstack.org/177078 | 05:26 |
| *** zhangtralon has joined #openstack-nova | 05:27 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: API: remove admin require from compute_node_get_all_by_* from db layer https://review.openstack.org/175334 | 05:28 |
| openstackgerrit | Eli Qiao proposed openstack/nova: API: Add policy enforcement test cases for pci API https://review.openstack.org/175745 | 05:28 |
| openstackgerrit | Eli Qiao proposed openstack/nova: API: remove admin require for compute_node(get_all/search_by_hyperviso) from db https://review.openstack.org/175744 | 05:28 |
| openstackgerrit | Eli Qiao proposed openstack/nova: API: remove admin require for compute_node_create/update/delete from db layer https://review.openstack.org/175693 | 05:28 |
| *** afazekas has quit IRC | 05:31 | |
| *** oro has joined #openstack-nova | 05:33 | |
| *** Nikolay_St has quit IRC | 05:33 | |
| *** Nikolay_St has joined #openstack-nova | 05:34 | |
| *** zhangtralon has quit IRC | 05:34 | |
| *** tojuvone has quit IRC | 05:35 | |
| *** Marga_ has quit IRC | 05:36 | |
| *** Marga_ has joined #openstack-nova | 05:36 | |
| *** Marga_ has quit IRC | 05:36 | |
| *** Marga_ has joined #openstack-nova | 05:37 | |
| *** arnaud___ has joined #openstack-nova | 05:39 | |
| *** VW_ has quit IRC | 05:41 | |
| *** Marga_ has quit IRC | 05:42 | |
| *** annashen has quit IRC | 05:42 | |
| *** annashen has joined #openstack-nova | 05:43 | |
| *** vilobhmm1 has quit IRC | 05:44 | |
| *** Nikolay_St has quit IRC | 05:48 | |
| *** jyuso1 has joined #openstack-nova | 05:49 | |
| *** afazekas has joined #openstack-nova | 05:49 | |
| *** ajayaa has joined #openstack-nova | 05:50 | |
| *** takedakn has joined #openstack-nova | 05:53 | |
| *** otter768 has joined #openstack-nova | 05:55 | |
| *** zhangtralon has joined #openstack-nova | 05:58 | |
| *** otter768 has quit IRC | 06:00 | |
| *** rushiagr_away is now known as rushiagr | 06:01 | |
| *** david-ly_ has joined #openstack-nova | 06:01 | |
| *** david-lyle has quit IRC | 06:03 | |
| *** josecastroleon has joined #openstack-nova | 06:03 | |
| openstackgerrit | Lan Qi Song proposed openstack/nova: Remove useless volume when boot from volume failed https://review.openstack.org/177084 | 06:04 |
| *** zhangtralon has quit IRC | 06:04 | |
| *** oro has quit IRC | 06:05 | |
| *** pkoniszewski has joined #openstack-nova | 06:08 | |
| *** cfriesen has quit IRC | 06:09 | |
| *** subscope_ has quit IRC | 06:11 | |
| *** annashen has quit IRC | 06:12 | |
| *** zhangtralon has joined #openstack-nova | 06:12 | |
| *** jwcroppe has joined #openstack-nova | 06:18 | |
| ekarlso | https://wiki.openstack.org/wiki/ReleaseNotes/Kilo#Known_Issues_2 < is that gonna be fixed for K ? | 06:19 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Imported Translations from Transifex https://review.openstack.org/177091 | 06:19 |
| *** unicell has quit IRC | 06:21 | |
| *** unicell has joined #openstack-nova | 06:23 | |
| *** rajesht has joined #openstack-nova | 06:29 | |
| *** arnaud___ has quit IRC | 06:33 | |
| *** lpetrut has joined #openstack-nova | 06:40 | |
| *** takedakn has quit IRC | 06:42 | |
| *** mpaolino has joined #openstack-nova | 06:42 | |
| *** salv-orlando has joined #openstack-nova | 06:45 | |
| *** sudipto has joined #openstack-nova | 06:45 | |
| *** romainh has joined #openstack-nova | 06:50 | |
| *** tojuvone has joined #openstack-nova | 06:51 | |
| *** armax has quit IRC | 06:51 | |
| *** hightall has joined #openstack-nova | 06:55 | |
| *** salv-orlando has quit IRC | 06:55 | |
| *** haomaiwang has quit IRC | 06:58 | |
| openstackgerrit | Rajesh Tailor proposed openstack/nova: Fix quota-update when nova-compute startup finish https://review.openstack.org/170118 | 07:00 |
| *** zhangtralon has quit IRC | 07:00 | |
| *** sudipto has quit IRC | 07:05 | |
| *** sahid has joined #openstack-nova | 07:06 | |
| *** coolsvap|afk is now known as coolsvap | 07:08 | |
| *** coolsvap is now known as coolsvap|afk | 07:09 | |
| *** salv-orlando has joined #openstack-nova | 07:12 | |
| *** eglynn has joined #openstack-nova | 07:12 | |
| *** pask81 has joined #openstack-nova | 07:12 | |
| *** takedakn has joined #openstack-nova | 07:12 | |
| *** salv-orlando has quit IRC | 07:14 | |
| *** sdake_ has quit IRC | 07:20 | |
| *** jichen has joined #openstack-nova | 07:21 | |
| openstackgerrit | IWAMOTO Toshihiro proposed openstack/nova: Omit RescheduledException in instance_fault.message https://review.openstack.org/165751 | 07:24 |
| *** achanda has quit IRC | 07:24 | |
| *** oro has joined #openstack-nova | 07:27 | |
| *** wuhg has quit IRC | 07:29 | |
| *** achanda has joined #openstack-nova | 07:36 | |
| *** daniel has joined #openstack-nova | 07:38 | |
| *** daniel is now known as Guest11440 | 07:39 | |
| *** salv-orlando has joined #openstack-nova | 07:40 | |
| *** Nikolay_St has joined #openstack-nova | 07:40 | |
| *** Guest11440 has quit IRC | 07:41 | |
| *** daniel_3 has joined #openstack-nova | 07:41 | |
| *** oro_ has joined #openstack-nova | 07:42 | |
| *** browne has quit IRC | 07:42 | |
| *** jistr has joined #openstack-nova | 07:44 | |
| *** jwcroppe has quit IRC | 07:44 | |
| *** tojuvone has quit IRC | 07:45 | |
| *** zhangtralon has joined #openstack-nova | 07:46 | |
| *** jwcroppe_ has joined #openstack-nova | 07:46 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Overwrite NovaException message https://review.openstack.org/177105 | 07:50 |
| *** kaisers1 has joined #openstack-nova | 07:50 | |
| *** alanf-mc has joined #openstack-nova | 07:51 | |
| *** zz_johnthetubagu is now known as johnthetubaguy | 07:51 | |
| *** jlanoux has joined #openstack-nova | 07:51 | |
| *** kaisers2 has joined #openstack-nova | 07:52 | |
| *** alanf-mc_ has joined #openstack-nova | 07:53 | |
| * johnthetubaguy makes big waking up noise | 07:53 | |
| *** lucasagomes has joined #openstack-nova | 07:54 | |
| openstackgerrit | Bartosz Fic proposed openstack/nova: Live migration failure in API doesn't set vm_state to Error https://review.openstack.org/168916 | 07:55 |
| *** alanf-mc has quit IRC | 07:56 | |
| *** otter768 has joined #openstack-nova | 07:56 | |
| *** tonytan4ever has quit IRC | 07:57 | |
| *** achanda has quit IRC | 07:57 | |
| *** jcoufal has joined #openstack-nova | 07:57 | |
| openstackgerrit | Sergey Nikitin proposed openstack/nova-specs: Implement server instance tagging https://review.openstack.org/177112 | 07:57 |
| *** kaisers2 has quit IRC | 07:59 | |
| *** alanf-mc_ has quit IRC | 08:00 | |
| *** otter768 has quit IRC | 08:01 | |
| *** atuvenie has joined #openstack-nova | 08:01 | |
| *** snikitin has joined #openstack-nova | 08:04 | |
| *** matrohon has joined #openstack-nova | 08:05 | |
| *** Mike_D_wk has quit IRC | 08:06 | |
| *** jistr is now known as jistr|mtg | 08:09 | |
| snikitin | johnthetubaguy: Hi, John! Now liberty is opened, so could you remove your -2, please? https://review.openstack.org/#/c/128940/ | 08:09 |
| johnthetubaguy | snikitin: let me check, normally involves getting your blueprint re-approved for liberty | 08:10 |
| johnthetubaguy | snikitin: yeah, can you please re-propose your spec for liberty, and add the Previously-Approved: kilo like in the commit please? | 08:10 |
| snikitin | johnthetubaguy: https://review.openstack.org/#/c/177112/ Done | 08:11 |
| johnthetubaguy | snikitin: superstar, thank you, let me take a look for you, would be good to get this one back up for review | 08:11 |
| *** redbeard has quit IRC | 08:12 | |
| johnthetubaguy | snikitin: there is a cross project spec on tagging I think, does this align with that? | 08:12 |
| johnthetubaguy | snikitin: here it is: https://review.openstack.org/#/c/155620/ | 08:13 |
| *** redbeard has joined #openstack-nova | 08:14 | |
| snikitin | johnthetubaguy: Hm, this is a new information for me. let me check | 08:16 |
| johnthetubaguy | snikitin: sorry, its new for me too, found out on Tuesday evening | 08:16 |
| *** jyuso1 has quit IRC | 08:19 | |
| openstackgerrit | Victor Stinner proposed openstack/nova-specs: Add "Port Nova to Python 3" spec https://review.openstack.org/176868 | 08:19 |
| openstackgerrit | Radoslav Gerganov proposed openstack/nova-specs: VMware: Native HTML5 consoles for VMware https://review.openstack.org/127283 | 08:20 |
| openstackgerrit | Takenori Yoshimatsu proposed openstack/nova: Catch AttributeError at instance.info_cache.delete https://review.openstack.org/167905 | 08:20 |
| *** yassine_ has joined #openstack-nova | 08:21 | |
| gilliard | Good morning Nova | 08:24 |
| *** gilliard is now known as gillliard | 08:24 | |
| johnthetubaguy | good morning | 08:25 |
| johnthetubaguy | or evening, depending on where you are | 08:25 |
| *** mpaolino has quit IRC | 08:26 | |
| *** andreykurilin__ has joined #openstack-nova | 08:27 | |
| bauzas | for that, we have the formula "mornooning" | 08:27 |
| johnthetubaguy | bauzas: somehow that sounds rude | 08:28 |
| gillliard | I'm pretty close to you, johnthetubaguy, globally speaking | 08:28 |
| gillliard | Same country, anyway | 08:28 |
| johnthetubaguy | gillliard: very true | 08:28 |
| johnthetubaguy | the second bit was for any Australians really | 08:29 |
| johnthetubaguy | I think I mentioned the empire once, but I got away with it | 08:30 |
| gillliard | ha | 08:30 |
| bauzas | johnthetubaguy: why ? morning/afternoon/evening :) | 08:30 |
| johnthetubaguy | bauzas: I dunno, spend too long in brass bands, they make most words sound rude somehow, with the correct level of winking and nudging | 08:31 |
| bauzas | johnthetubaguy: eh should I mention I'm French re: your empire ? :p | 08:31 |
| johnthetubaguy | bauzas: :) | 08:31 |
| bauzas | johnthetubaguy: ack, I'm missing some semantics here as that's not my native language :) | 08:32 |
| bauzas | oooooh Friday ! | 08:32 |
| *** bauzas is now known as bauwser | 08:32 | |
| johnthetubaguy | lol | 08:32 |
| johnthetubaguy | snikitin: hi, just had a look at the code too, your code is closer than the spec I think | 08:33 |
| johnthetubaguy | snikitin: the bit we might want to be clear on is the filtering, I am OK us adding that later, if that helps us move forward? | 08:33 |
| johnthetubaguy | snikitin: I think we should be able to merge more API patches in liberty, now v2.1 is in better shape, FWIW, lets try get this merged before the summit! | 08:34 |
| *** derekh has joined #openstack-nova | 08:35 | |
| *** pixelb has joined #openstack-nova | 08:37 | |
| *** ndipanov has joined #openstack-nova | 08:37 | |
| *** ndipanov is now known as ndipanoff | 08:38 | |
| snikitin | johnthetubaguy: Great, let's try! So as I understand, I need to modify my spec to syncronise it with that big spec, right? | 08:38 |
| *** aix has joined #openstack-nova | 08:40 | |
| johnthetubaguy | snikitin: yeah, its mostly what you have done in your code already | 08:40 |
| johnthetubaguy | snikitin: feel free to not implement some bits, like the filtering, I think its just when things are present, they should be the same (at least thats my thinking here), we can add the filtering bits later | 08:41 |
| johnthetubaguy | snikitin: your spec doesn't mention adding tags into the server details response, like your code actually does. its worth adding that into the spec too | 08:42 |
| johnthetubaguy | snikitin: its not mentioned in the big spec, but thats fine | 08:42 |
| johnthetubaguy | alex_xu: hi, just about to revamp the relax validation spec | 08:42 |
| *** salv-orlando has quit IRC | 08:43 | |
| snikitin | johnthetubaguy: Ok, thank you for new information! I'll do my best. | 08:44 |
| johnthetubaguy | snikitin: appreciate your hard work on this, sorry for all the changes, but at least its now and not as we were about to press approve on the code! | 08:49 |
| snikitin | johnthetubaguy: No problems :) | 08:50 |
| *** TobiasE has joined #openstack-nova | 08:53 | |
| alex_xu | johnthetubaguy: hi, I saw you replied | 08:56 |
| alex_xu | johnthetubaguy: how about add some description about why relax validation is more smooth for operator switch to v2.1? That is good for persuade people? | 08:57 |
| johnthetubaguy | alex_xu: yeah, thats a good idea | 08:58 |
| johnthetubaguy | alex_xu: the other thing is describing how I expect clients to update so they get the validation | 08:58 |
| alex_xu | johnthetubaguy: +1 | 08:59 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/176844 | 09:01 |
| *** tojuvone has joined #openstack-nova | 09:02 | |
| *** ndipanoff has quit IRC | 09:02 | |
| *** ndipanov has joined #openstack-nova | 09:10 | |
| bauwser | alex_xu: I was on day-off yesterday so I missed your discussions about the non-admin context check | 09:11 |
| bauwser | alex_xu: basically, what's the outcome ? | 09:11 |
| bauwser | johnthetubaguy: I also have a question for you about your opinions on a change for a python client thing | 09:12 |
| johnthetubaguy | bauwser: fire away | 09:12 |
| bauwser | johnthetubaguy: so, some upstream effort is done outside Nova for automatically evacuating hosts based on Pacemaker | 09:12 |
| *** annashen has joined #openstack-nova | 09:13 | |
| bauwser | johnthetubaguy: within those scripts, they call the CLI | 09:13 |
| *** ndipanov is now known as ndipanoff | 09:13 | |
| bauwser | johnthetubaguy: and in particular nova service-is-up | 09:13 |
| alex_xu | bauwser: there is patch for it in master https://review.openstack.org/#/c/176651/, and the https://review.openstack.org/160089 stop to merge, after 176651 merged, we can back port it first, is it ok for you? | 09:13 |
| bauwser | johnthetubaguy: now the point is, nova service-is-up is returning 0 even if the service is dowb | 09:13 |
| bauwser | down even | 09:13 |
| bauwser | johnthetubaguy: because the contract was OK -ie the API returned 200 or something else | 09:14 |
| bauwser | johnthetubaguy: you got my point ? why couldn't we assume that nova service-is-up should return anything but zero if the service is down | 09:15 |
| bauwser | johnthetubaguy: and if so, shouldn't we assume another HTTP code for the API side ? | 09:15 |
| alex_xu | bauwser: we only have index method for service, that is list all the serivce status. the HTTP code doesn't works. | 09:16 |
| bauwser | alex_xu: okay, I have to dig into the client side then | 09:17 |
| bauwser | alex_xu: so I assume it's not a direct translation of the API result body, but rather a calculation on the CLI side ? | 09:17 |
| *** annashen has quit IRC | 09:17 | |
| * bauwser needs to jump into the code | 09:17 | |
| alex_xu | bauwser: yea, I guess so, maybe can do something in the client, not the server side | 09:18 |
| bauwser | alex_xu: my take is that it's really clear that by default, the CLI is returning zero if the command succeeded, 1 the other way | 09:18 |
| johnthetubaguy | bauwser: alex_xu: is this something we broke, or a bug we are trying to fix for them? | 09:19 |
| johnthetubaguy | its tempting to add a spec for this change, so we know how its going to be used | 09:19 |
| bauwser | johnthetubaguy: I don't have all the context, my take is to discuss that correctly | 09:19 |
| bauwser | johnthetubaguy: I would tend to say I agree with you, but I just wonder how all of this is just no-brainer | 09:20 |
| bauwser | alex_xu: I need to check what does the CLI actually | 09:20 |
| bauwser | I mean the pythonclient API thing | 09:20 |
| bauwser | not the shell of course | 09:20 |
| alex_xu | bauwser: yea I guess | 09:21 |
| bauwser | mmmm | 09:22 |
| alex_xu | bauwser: you want to make the cli can be used by script. that sounds make sense to me also. but need take a look at how to implement that. | 09:25 |
| bauwser | alex_xu: no you misunderstood me | 09:25 |
| bauwser | alex_xu: the CLI (tbp the "shell" module) is just a wrapper on top of the novaclient API | 09:26 |
| bauwser | alex_xu: there is no notion of return codes in the client API | 09:26 |
| bauwser | alex_xu: but rather client exceptions | 09:26 |
| alex_xu | bauwser: emm....I need dig into more client code also :) | 09:27 |
| bauwser | so, the point is : should the API counterpart of 'nova service-is-up' just raise an Exception if the service is not up ? | 09:27 |
| bauwser | alex_xu: https://github.com/openstack/python-novaclient/blob/master/novaclient/shell.py#L905-L916 | 09:27 |
| *** hightall has quit IRC | 09:27 | |
| bauwser | alex_xu: that's how retcodes are generated | 09:27 |
| openstackgerrit | jichenjc proposed openstack/nova: Create Nova Scheduler Num Instances Weighter https://review.openstack.org/130526 | 09:28 |
| *** jyuso1 has joined #openstack-nova | 09:28 | |
| alex_xu | bauwser: ok | 09:28 |
| bauwser | alex_xu: because you can use the client directly in other projects | 09:28 |
| bauwser | alex_xu: you just import it and do the auth | 09:29 |
| alex_xu | bauwser: yes | 09:29 |
| alex_xu | bauwser: sounds make sense | 09:29 |
| *** jistr|mtg is now known as jistr | 09:32 | |
| bauwser | alex_xu: I actually played a lot with the novaclient in a previous life :) | 09:33 |
| *** salv-orlando has joined #openstack-nova | 09:35 | |
| alex_xu | bauwser: a previous life means... | 09:36 |
| bauwser | alex_xu: means when I was fool enough to be an operator writing python scripts - and then working upstream on a separate project called Climate | 09:37 |
| *** sdake has joined #openstack-nova | 09:37 | |
| alex_xu | bauwser: ah, Blazar | 09:37 |
| bauwser | yup | 09:37 |
| alex_xu | bauwser: I remember you said there are Blazar may move into gantt discussion at summit? | 09:38 |
| bauwser | alex_xu: tl;dr: Blazar is defunct | 09:38 |
| alex_xu | bauwser: :), expect re-born | 09:39 |
| *** sdake_ has joined #openstack-nova | 09:39 | |
| *** lpetrut has quit IRC | 09:39 | |
| *** ankit_ag has joined #openstack-nova | 09:40 | |
| bauwser | alex_xu: not a good idea now :) | 09:40 |
| bauwser | alex_xu: that's not the best idea I had | 09:40 |
| bauwser | :) | 09:40 |
| alex_xu | bauwser: ok :) | 09:41 |
| *** jwcroppe_ has quit IRC | 09:41 | |
| *** jwcroppe has joined #openstack-nova | 09:42 | |
| *** haomaiwang has joined #openstack-nova | 09:42 | |
| *** sdake has quit IRC | 09:43 | |
| *** abhishekk has joined #openstack-nova | 09:45 | |
| ankit_ag | Hi all, I am getting following error while running run_tests.sh with code coverage | 09:45 |
| ankit_ag | RuntimeError: No 'nova.ipv6_backend' driver found, looking for 'rfc2462' | 09:45 |
| ankit_ag | does someone faced the same issue | 09:45 |
| bauwser | ankit_ag: try rebuilding your venv | 09:45 |
| bauwser | ankit_ag: ./runtests.sh -y | 09:46 |
| bauwser | oops | 09:46 |
| bauwser | ./run_tests.sh -u | 09:46 |
| ankit_ag | bauwser: I has tried it with -f | 09:46 |
| bauwser | ankit_ag: oh ? | 09:46 |
| ankit_ag | nut no luck | 09:46 |
| ankit_ag | but no luck | 09:47 |
| bauwser | IIRC, that just means that the entrypoints haven't been correctly recreated | 09:48 |
| *** Marga_ has joined #openstack-nova | 09:48 | |
| bauwser | ankit_ag: oh you said with code coverage ? | 09:49 |
| ankit_ag | bauwser: yes with code coverage it gives error, while I am able to run test suit successfully without code coverage | 09:50 |
| *** Marga_ has quit IRC | 09:50 | |
| bauwser | ankit_ag: ok, lemme look | 09:50 |
| *** Marga_ has joined #openstack-nova | 09:50 | |
| ankit_ag | bauwser: okey | 09:51 |
| *** Longgeek_ has joined #openstack-nova | 09:52 | |
| *** Longgeek has quit IRC | 09:52 | |
| bauwser | ankit_ag: I just wonder if that already worked... | 09:52 |
| gillliard | ankit_ag: I am running the coverage job daily, haven't noticed any failures. | 09:52 |
| gillliard | I do: "tox -ecover" | 09:53 |
| ankit_ag | gillliard: I am trying it with ./run_tests.sh --coverage | 09:54 |
| bauwser | honestly, I don't see the point of checking nova coverage :) | 09:55 |
| ankit_ag | gillliard: also I have done fresh installation using devstack 2-3 times, but still facing the same issue | 09:55 |
| bauwser | just because there are *lots* of tests which are performing deeply up to the DB and are not isolated | 09:56 |
| bauwser | so there are high changes that a report should mention a code section to be "covered" while it's just checking its nominal state | 09:56 |
| gillliard | agreed, bauwser, | 09:57 |
| ankit_ag | bauwser: yes you are right | 09:57 |
| *** otter768 has joined #openstack-nova | 09:57 | |
| ankit_ag | bauwser: i just wanted to check coverage for the code I have written | 09:57 |
| bauwser | I saw boris-42's email, I should have replied - just because Nova doesn't work that way unfortunately :( | 09:58 |
| bauwser | ankit_ag: honestly, you won't get more confidence | 09:58 |
| openstackgerrit | jichenjc proposed openstack/nova: Skip only one host weight calculation https://review.openstack.org/137275 | 09:58 |
| ankit_ag | bauwser, gillliard: I will try this using tox as mentioned by gillliard. | 09:58 |
| bauwser | ankit_ag: but by reading the runtests stuff, I just want to say to use like like gillliard says | 09:58 |
| openstackgerrit | Eli Qiao proposed openstack/nova: WIP: Add ConsoleAuthToken object and console_auth_token table https://review.openstack.org/177078 | 09:59 |
| openstackgerrit | Eli Qiao proposed openstack/nova: WIP Conductor: Add console_auth_token apis https://review.openstack.org/177144 | 09:59 |
| openstackgerrit | Eli Qiao proposed openstack/nova: WIP Allow running multiple consoleauth services https://review.openstack.org/177145 | 09:59 |
| ankit_ag | bauwser, gillliard: yes , thanks for your time and for looking in to the issue :) | 09:59 |
| gillliard | I can't remember why I do it that way - I remember it being a PITA to get it working. Several openstack projects had completely broken coverage tests, because (I assume) it's not actually run during any of the CI. | 10:00 |
| gillliard | Perhaps we should just remove it ;) ? | 10:01 |
| *** zhenguo has quit IRC | 10:02 | |
| *** otter768 has quit IRC | 10:03 | |
| boris-42 | bauwser: hm?) | 10:04 |
| boris-42 | bauwser: we have few fixes of that job | 10:05 |
| boris-42 | bauwser: finally version is here https://github.com/openstack/rally/blob/master/tests/ci/cover.sh | 10:05 |
| *** flaper87 has quit IRC | 10:05 | |
| *** flaper87 has joined #openstack-nova | 10:05 | |
| *** markvoelker has joined #openstack-nova | 10:06 | |
| *** Marga_ has quit IRC | 10:06 | |
| boris-42 | gillliard: perhaps it is better to fix thing? | 10:06 |
| openstackgerrit | Tomi Juvonen proposed openstack/nova-specs: New nova API call to mark nova-compute down https://review.openstack.org/169836 | 10:06 |
| boris-42 | gillliard: IMHO coverage is very important in case of python | 10:06 |
| boris-42 | gillliard: like in case of Rally it helped a lot to port it to python34 | 10:07 |
| gillliard | boris-42: OK I don't know much about Rally, I admit. But the coverage-level of nova is a little difficult to interpret, for the reasons bauwser mentioned a few minutes ago. | 10:07 |
| gillliard | If it's providing some value, somewhere, then by all means we should fix it :) | 10:08 |
| *** markvoelker has quit IRC | 10:11 | |
| *** jwcroppe has quit IRC | 10:12 | |
| openstackgerrit | Tomi Juvonen proposed openstack/nova-specs: New nova API call to mark nova-compute down https://review.openstack.org/169836 | 10:15 |
| boris-42 | gillliard: it has a lot of value | 10:16 |
| boris-42 | gillliard: if it is used and used properly | 10:17 |
| openstackgerrit | jichenjc proposed openstack/nova: libvirt: catch libvirtError when init_host https://review.openstack.org/177151 | 10:19 |
| *** Bsony has joined #openstack-nova | 10:20 | |
| *** strictlyb is now known as sb | 10:21 | |
| gillliard | boris-42: bauwser: BTW I have been keeping this to monitor nova coverage http://15.126.204.92:9000/drilldown/measures/1?metric=coverage | 10:22 |
| *** IanGovett has joined #openstack-nova | 10:22 | |
| *** Longgeek_ has quit IRC | 10:22 | |
| *** Longgeek has joined #openstack-nova | 10:23 | |
| *** andreykurilin__ has quit IRC | 10:26 | |
| johnthetubaguy | dansmith: just wanted to check what bug your note on evacuate matches: https://wiki.openstack.org/w/index.php?title=ReleaseNotes/Kilo&diff=prev&oldid=75612 | 10:26 |
| johnthetubaguy | dansmith: wondering if we could just default to not starting, rather than deleting VMs, although its super late for that now | 10:26 |
| boris-42 | gillliard: it is useless to monitor coverage | 10:27 |
| boris-42 | gillliard: job should be just blocking and they you don't need to take care | 10:27 |
| boris-42 | gillliard: about monitoring | 10:28 |
| gillliard | boris-42: what do you mean, blocking? | 10:29 |
| gillliard | boris-42: as in, CI would fail if your patch decreases coverage? | 10:30 |
| boris-42 | gillliard: in rally we are running coverage job on every check and if patch adds more then X (in our case 4 lines) without tests it fails | 10:30 |
| boris-42 | gillliard: yep | 10:30 |
| *** tojuvone has quit IRC | 10:32 | |
| *** garyk has joined #openstack-nova | 10:33 | |
| gillliard | boris-42: In my experience, strict coverage %age requirements don't correlate well with good testing. | 10:33 |
| gillliard | I just use the coverage tool for exploration of sparsely-tested areas of the code. | 10:34 |
| gillliard | I guess, if it helps keep reviewers focussed on tests then that's a good thing, though. | 10:35 |
| *** samueldmq has joined #openstack-nova | 10:36 | |
| samueldmq | good morning :) | 10:36 |
| boris-42 | gillliard: I calculated that it safes about 25 hours of my reviewing time during single release | 10:37 |
| *** vladikr has quit IRC | 10:37 | |
| boris-42 | gillliard: and I am happy | 10:37 |
| gillliard | wow. I'd be happy too :) | 10:38 |
| boris-42 | gillliard: because for example I did 730 reviews / release | 10:39 |
| boris-42 | gillliard: and every thing that can be automated helps | 10:40 |
| boris-42 | gillliard: btw I agree that just 100% coverage doesn't mean too much | 10:40 |
| boris-42 | gillliard: but I just don't want to spend time on patch that hasn't 100% coverage at all (because it won't be merged in any case) | 10:40 |
| gillliard | boris-42: Uh-huh. Using it as a filter to decide what to review makes a lot more sense than using it as a substitute for reviews. Most projects I've worked on would merge anything that passed CI | 10:43 |
| boris-42 | gillliard: sometimes the same happens in Rally unforutnatelly=( | 10:44 |
| boris-42 | gillliard: but at least no one criteria will be checked by Ci=) | 10:44 |
| openstackgerrit | jichenjc proposed openstack/nova: Create Nova Scheduler Num Instances Weighter https://review.openstack.org/130526 | 10:51 |
| *** sdake_ has quit IRC | 10:53 | |
| *** zhangtralon has quit IRC | 10:59 | |
| openstackgerrit | sahid proposed openstack/nova: libvirt: use importutils instead of python built-in https://review.openstack.org/176372 | 11:04 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: move destroy code in guest https://review.openstack.org/174315 | 11:04 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: move get domain info in guest https://review.openstack.org/174314 | 11:04 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduces a new Guest object https://review.openstack.org/174313 | 11:04 |
| *** markvoelker has joined #openstack-nova | 11:07 | |
| *** bfic has joined #openstack-nova | 11:07 | |
| *** lpetrut has joined #openstack-nova | 11:08 | |
| *** markvoelker has quit IRC | 11:12 | |
| samueldmq | hi, using a domain-scoped token or an unscoped token for an API call, nova says | 11:14 |
| samueldmq | 400 Malformed request URL: URL's project_id 'abbe4eac077b42efa5f7872925f10d93' doesn't match Context's project_id 'None' | 11:14 |
| samueldmq | I think UX could be improved by raising a 401 unauthorized, what do you think about ? | 11:15 |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: Removed unused variables, methods and classes https://review.openstack.org/164666 | 11:16 |
| *** alexpilotti has joined #openstack-nova | 11:18 | |
| ekarlso | so if you change names in kilo nova will delete stuff ? :p | 11:18 |
| samueldmq | ekarlso, are you replying me ? if so I don't follow what you're saying :) | 11:20 |
| *** wznoinsk has quit IRC | 11:22 | |
| *** dguitarbite has quit IRC | 11:24 | |
| ekarlso | samueldmq: nope ;p | 11:25 |
| samueldmq | ekarlso, :-) | 11:25 |
| *** dguitarbite has joined #openstack-nova | 11:25 | |
| *** yamahata has quit IRC | 11:31 | |
| *** TobiasE has quit IRC | 11:31 | |
| *** bfic has quit IRC | 11:33 | |
| openstackgerrit | Hans Lindgren proposed openstack/nova: Merge security group client side rpc api with compute rpc api https://review.openstack.org/177186 | 11:36 |
| openstackgerrit | Victor Stinner proposed openstack/nova-specs: Add "Port Nova to Python 3" spec https://review.openstack.org/176868 | 11:38 |
| *** Marga_ has joined #openstack-nova | 11:41 | |
| *** lucasagomes is now known as lucas-hungry | 11:45 | |
| *** dprince has joined #openstack-nova | 11:49 | |
| *** dguitarbite has quit IRC | 11:53 | |
| *** markvoelker has joined #openstack-nova | 11:54 | |
| *** atuvenie has quit IRC | 11:56 | |
| *** Marga_ has quit IRC | 11:57 | |
| *** diegows has joined #openstack-nova | 11:57 | |
| *** otter768 has joined #openstack-nova | 11:58 | |
| *** ganso_ has joined #openstack-nova | 11:59 | |
| *** snikitin has quit IRC | 11:59 | |
| *** zhenguo has joined #openstack-nova | 11:59 | |
| *** EmilienM has quit IRC | 12:01 | |
| *** EmilienM has joined #openstack-nova | 12:01 | |
| *** bfic has joined #openstack-nova | 12:02 | |
| *** otter768 has quit IRC | 12:02 | |
| *** exploreshaifali has joined #openstack-nova | 12:04 | |
| *** dboik_ has quit IRC | 12:10 | |
| *** subscope has quit IRC | 12:11 | |
| *** erkules_ is now known as erkules | 12:11 | |
| *** erkules has joined #openstack-nova | 12:11 | |
| *** mpaolino has joined #openstack-nova | 12:11 | |
| *** snikitin has joined #openstack-nova | 12:14 | |
| *** pkoniszewski has quit IRC | 12:15 | |
| *** IanGovett1 has joined #openstack-nova | 12:17 | |
| *** IanGovett has quit IRC | 12:17 | |
| openstackgerrit | Divya K Konoor proposed openstack/nova: view hypervisor details rest api should be allowed for non-admins https://review.openstack.org/176651 | 12:20 |
| *** foexle has joined #openstack-nova | 12:20 | |
| *** oro has quit IRC | 12:22 | |
| *** oro_ has quit IRC | 12:23 | |
| *** dave-mccowan has joined #openstack-nova | 12:24 | |
| *** subscope has joined #openstack-nova | 12:25 | |
| *** dboik has joined #openstack-nova | 12:27 | |
| *** dboik has quit IRC | 12:27 | |
| *** baoli has joined #openstack-nova | 12:27 | |
| *** dboik has joined #openstack-nova | 12:27 | |
| *** VW_ has joined #openstack-nova | 12:27 | |
| *** snikitin has quit IRC | 12:29 | |
| *** ivasev has joined #openstack-nova | 12:29 | |
| *** popw1 has quit IRC | 12:30 | |
| openstackgerrit | Petrut Lucian proposed openstack/nova-specs: Hyper-V: Add storage QoS support https://review.openstack.org/175422 | 12:30 |
| *** joefides has quit IRC | 12:30 | |
| *** dboik_ has joined #openstack-nova | 12:35 | |
| *** baoli has quit IRC | 12:35 | |
| *** dboik_ has quit IRC | 12:35 | |
| *** dboik_ has joined #openstack-nova | 12:36 | |
| *** snikitin has joined #openstack-nova | 12:36 | |
| *** dguitarbite has joined #openstack-nova | 12:36 | |
| openstackgerrit | jichenjc proposed openstack/nova: Add os_shutdown_timeout support https://review.openstack.org/177217 | 12:37 |
| *** dboik has quit IRC | 12:38 | |
| *** raildo has joined #openstack-nova | 12:39 | |
| *** baoli has joined #openstack-nova | 12:40 | |
| *** edleafe is now known as figleaf | 12:41 | |
| openstackgerrit | Bartosz Fic proposed openstack/nova: Live migration failure in API doesn't set vm_state to Error https://review.openstack.org/168916 | 12:42 |
| openstackgerrit | Eli Qiao proposed openstack/nova: Overwrite NovaException message https://review.openstack.org/177105 | 12:43 |
| eliqiao | gate's broken again? why I see Internal Server Error ? http://logs.openstack.org/78/177078/2/check/gate-nova-python27/e8fff4f/console.html | 12:45 |
| *** lucas-hungry is now known as lucasagomes | 12:46 | |
| *** diegows has quit IRC | 12:46 | |
| *** busterswt has joined #openstack-nova | 12:47 | |
| gillliard | eliqiao: It's been noticed in #openstack-infra | 12:48 |
| *** redbeard has quit IRC | 12:49 | |
| openstackgerrit | sahid proposed openstack/nova: libvirt: add get_hardware_info method to Host https://review.openstack.org/177222 | 12:49 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: add get_cpu_stats method to host https://review.openstack.org/177223 | 12:49 |
| *** xyang1 has joined #openstack-nova | 12:50 | |
| sdague | eliqiao: it's just the log display tool, I'm starting to unwind some landed patches | 12:50 |
| *** mwagner_lap has quit IRC | 12:51 | |
| *** ajayaa has quit IRC | 12:51 | |
| openstackgerrit | Hans Lindgren proposed openstack/nova: Add security group calls missing from latest compute rpc api version bump https://review.openstack.org/177186 | 12:52 |
| bauwser | johnthetubaguy: have a few mins for discussing about https://review.openstack.org/#/c/174764/4/specs/liberty/approved/platform-health-aware-scheduling.rst,cm ? | 12:52 |
| bauwser | johnthetubaguy: because I think we need to be clear on what is a monitor, and if we want to keep them | 12:52 |
| bauwser | johnthetubaguy: I also noticed a separate patch using those, that's why I think we need to state a guidance over here | 12:53 |
| bauwser | johnthetubaguy: ie. do we allow extensibility, and if so, what's acceptable ? | 12:53 |
| *** afazekas has quit IRC | 12:55 | |
| *** dboik_ has quit IRC | 12:55 | |
| *** dboik has joined #openstack-nova | 12:56 | |
| *** yassine_ has quit IRC | 12:56 | |
| *** dboik has quit IRC | 12:56 | |
| *** dboik has joined #openstack-nova | 12:57 | |
| *** VW_ has quit IRC | 12:57 | |
| *** alaski is now known as lascii | 12:58 | |
| *** joefides has joined #openstack-nova | 12:59 | |
| *** yassine has joined #openstack-nova | 12:59 | |
| *** signed8bit has joined #openstack-nova | 13:00 | |
| *** signed8bit has quit IRC | 13:00 | |
| johnthetubaguy | bauwser: sorry, was having lunch, but you are right, we should document guidance some where, probably devref, I need to start writing those, I have a list in my tasks | 13:02 |
| *** daniel_3 has quit IRC | 13:02 | |
| *** nkrinner has quit IRC | 13:03 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/176844 | 13:03 |
| bauwser | johnthetubaguy: I left a few comments on https://review.openstack.org/#/c/174764/4 | 13:04 |
| bauwser | johnthetubaguy: I think you can take some of those if you agree for some base on your devref | 13:05 |
| *** neelashah has joined #openstack-nova | 13:05 | |
| bauwser | johnthetubaguy: ie. no 3rd party tool, just hypervisor-driven | 13:05 |
| bauwser | johnthetubaguy: needs to be gated using the regular CI or the existing driver CIs | 13:05 |
| *** eglynn has quit IRC | 13:06 | |
| bauwser | johnthetubaguy: and I probably forget many others... | 13:06 |
| *** artom has joined #openstack-nova | 13:06 | |
| *** oro has joined #openstack-nova | 13:07 | |
| bauwser | johnthetubaguy: but in some way, we need to keep a certain extensibility offered by the monitors, just because the scheduler allows that | 13:07 |
| *** IanGovett1 has quit IRC | 13:07 | |
| johnthetubaguy | bauwser: so part of me actually likes an API to let third party tools send data to our scheduler, but it should be very clearly external, and have a nice strong API, I don't want us to do the gathering | 13:07 |
| *** IanGovett has joined #openstack-nova | 13:07 | |
| *** ajayaa has joined #openstack-nova | 13:07 | |
| bauwser | johnthetubaguy: yeah we need to be gentle and offer some out-of-tree possibility | 13:08 |
| bauwser | johnthetubaguy: but on the other hand, that's just something which needs to stay out-of-tree | 13:08 |
| johnthetubaguy | bauwser: yeah, I think we agree | 13:09 |
| bauwser | johnthetubaguy: ie. I don't care if some vendors are using the monitors to gather metrics from anything else but the hypervisor | 13:09 |
| bauwser | johnthetubaguy: and if they deliver some out-of-tree filters using the configurability offered | 13:09 |
| bauwser | johnthetubaguy: but all of that is out-of-tree | 13:09 |
| johnthetubaguy | OK, thats stricter than what I was thinking | 13:09 |
| bauwser | johnthetubaguy: what becomes in-tree needs CI | 13:09 |
| johnthetubaguy | but actually, that makes more sense in a way | 13:10 |
| openstackgerrit | Ph. Marek proposed openstack/nova-specs: Connecting Nova to DRBD storage nodes directly. https://review.openstack.org/134153 | 13:10 |
| bauwser | johnthetubaguy: the problem is, say that we offer an external API like the hypervisor one | 13:10 |
| bauwser | johnthetubaguy: how can we get pre-merge on that ? | 13:10 |
| bauwser | gate$ | 13:10 |
| bauwser | johnthetubaguy: the PCI stuff is just a pain we have | 13:11 |
| johnthetubaguy | so with whats in my head, its testable in gate | 13:11 |
| bauwser | johnthetubaguy: it took like 3 cycles to get a CI | 13:11 |
| johnthetubaguy | totally | 13:11 |
| *** IanGovett has quit IRC | 13:11 | |
| bauwser | johnthetubaguy: I just don't want to reproduce that using the monitors | 13:11 |
| johnthetubaguy | comes back to the extensible resource tracker stuff again, thinking about it | 13:11 |
| johnthetubaguy | bauwser: yeah, agreed | 13:11 |
| bauwser | johnthetubaguy: and btw. I would like to shoot one day the AttestationFilter - or gate it | 13:12 |
| johnthetubaguy | bauwser: those filters don't worry me too much, if you get the unit tests right | 13:12 |
| johnthetubaguy | bauwser: the PCI stuff is different, that is spread all over the shop | 13:12 |
| *** abhishekk has quit IRC | 13:12 | |
| sdague | so, isn't this really the next step for the scheduler anyways. Openning up a REST API that lets more than nova feed it data? | 13:12 |
| bauwser | johnthetubaguy: I should just point you out some internal bugs I have with that filter :) | 13:13 |
| *** afazekas has joined #openstack-nova | 13:13 | |
| bauwser | sdague: the problem is not in the future | 13:13 |
| johnthetubaguy | sdague: yeah, it does feel in the right direction | 13:13 |
| bauwser | sdague: the problem is now | 13:13 |
| bauwser | sdague: but I agree, we'll opening up the API | 13:13 |
| sdague | bauwser: so, I realize we have a lot of now problems | 13:13 |
| bauwser | sdague: like we did using the client and the update_resource_stats thign | 13:13 |
| sdague | yep | 13:13 |
| bauwser | sdague: but then, we need to offer a contract | 13:14 |
| *** oro has quit IRC | 13:14 | |
| bauwser | sdague: because we also claim against those resources | 13:14 |
| sdague | bauwser: sure, but that was always going to have to happen | 13:14 |
| sdague | if the scheduler was ever going to be more than just nova, it needs an externalized contract | 13:14 |
| bauwser | sdague: so the monitors are only a small and nice abstraction for something which doesn't need to be claimed | 13:15 |
| bauwser | sdague: that's what we're trying to do using the scheduler client and the clean-up stuff | 13:15 |
| bauwser | sdague: take the bp/resource-usage as a reference | 13:15 |
| *** eharney has joined #openstack-nova | 13:15 | |
| bauwser | sdague: but coming back to the monitors, those are only metrics | 13:16 |
| bauwser | sdague: if we're opening up the gates and leave anything come to the scheduler, then we'll suffer the same issues as the PCI and NUMA guys had | 13:16 |
| bauwser | sdague: so my point is - yes, we're opening up by some extend what can be reported to Nova | 13:17 |
| bauwser | and to the scheduler | 13:17 |
| sdague | so, I don't understand how anyone imagined the scheduler as an independent project without that ability | 13:17 |
| *** sb has quit IRC | 13:17 | |
| bauwser | which ability ? | 13:17 |
| *** strictlyb has joined #openstack-nova | 13:17 | |
| bauwser | to extend the API ? | 13:18 |
| bauwser | or to extend what can be reported ? | 13:18 |
| *** kiran-r has quit IRC | 13:18 | |
| sdague | what's reported and used for scheduling | 13:19 |
| bauwser | sdague: again, we're thinking of that and that's in track | 13:19 |
| *** vladikr has joined #openstack-nova | 13:19 | |
| *** rushil has joined #openstack-nova | 13:19 | |
| bauwser | sdague: http://specs.openstack.org/openstack/nova-specs/specs/kilo/approved/resource-objects.html | 13:19 |
| sdague | yeh, so I'd just punt on this and say - this requires the contract for such kinds of things coming in for nova scheduler | 13:19 |
| bauwser | jaypipes just needs to update that spec and resubmit it | 13:20 |
| bauwser | jaypipes: that's a call to you :p | 13:20 |
| *** rushil has quit IRC | 13:20 | |
| *** VW_ has joined #openstack-nova | 13:20 | |
| bauwser | sdague: once that bp will be merged, we'll be having a contract between nova and the scheduler | 13:21 |
| *** rushil has joined #openstack-nova | 13:21 | |
| *** VW_ has quit IRC | 13:21 | |
| *** mpaolino has quit IRC | 13:22 | |
| *** mpaolino has joined #openstack-nova | 13:23 | |
| *** zz_jgrimm is now known as jgrimm | 13:23 | |
| sdague | right, but still not an externalized contract for the scheduler | 13:24 |
| sdague | it seems like conceptually you need | 13:25 |
| sdague | nova <- internal contract -> scheduler | 13:25 |
| *** jgrimm is now known as zz_jgrimm | 13:25 | |
| sdague | nova <- internal contract -> scheduler <- external contract -> [stuff] | 13:25 |
| sdague | nova + [stuff] <- external contract -> scheduler | 13:26 |
| sdague | then you can move the schedule into a separate project (if that still makes sense) | 13:26 |
| *** artom has quit IRC | 13:27 | |
| *** artom_ has joined #openstack-nova | 13:27 | |
| bauwser | sdague: agreed but then comes the problem of the scheduler scalability | 13:27 |
| *** xyang1 has quit IRC | 13:27 | |
| bauwser | sdague: because by design, the Scheduler API takes no writes | 13:27 |
| *** xyang1 has joined #openstack-nova | 13:28 | |
| sdague | yeh, so that's not going to work if you want it out of tree :) | 13:28 |
| bauwser | sdague: that's why our mediator is the conductor | 13:28 |
| *** kaisers1 has quit IRC | 13:28 | |
| bauwser | sdague: agreed, I never spoke about a separate project :) | 13:28 |
| bauwser | at least atm | 13:28 |
| sdague | sure | 13:28 |
| *** kaisers has quit IRC | 13:28 | |
| *** cbader has joined #openstack-nova | 13:29 | |
| bauwser | yeah the Gantt naming made lots of confusion that I need to sort out | 13:29 |
| *** eglynn has joined #openstack-nova | 13:29 | |
| *** kaisers has joined #openstack-nova | 13:30 | |
| bauwser | so, basically, we need to figure out how to provide stats to the scheduler - not using the conductor - and my take is that it needs some changes | 13:30 |
| *** kaisers1 has joined #openstack-nova | 13:30 | |
| *** bfic has quit IRC | 13:30 | |
| sdague | yep, sounds right | 13:30 |
| bauwser | I'm unsure it can be faced for Liberty | 13:30 |
| bauwser | s/unsure/really not sure | 13:31 |
| sdague | yep, that's fine, honestly, rather would do it right than do it fast | 13:31 |
| johnthetubaguy | sdague: +1 that statement here | 13:32 |
| *** diegows has joined #openstack-nova | 13:32 | |
| johnthetubaguy | bauwser: I am thinking we have a REST API so you can set some statistic about a host, key/value and expiry, and enable accepting updates in batches, or something like that | 13:32 |
| *** Nikolay_St has quit IRC | 13:33 | |
| bauwser | johnthetubaguy: eh, you mean the /os-hypervisors ? | 13:33 |
| bauwser | because /os-hosts is just a wrong thing | 13:33 |
| *** erlon has joined #openstack-nova | 13:33 | |
| bauwser | and because /os-services is just service related :) | 13:33 |
| bauwser | johnthetubaguy: oh, you mean if we would have an endpoint for that ? | 13:34 |
| johnthetubaguy | bauwser: it would be new, start again thing | 13:34 |
| bauwser | johnthetubaguy: I think the problem is not how to get things | 13:34 |
| bauwser | johnthetubaguy: but rather how to claim them :) | 13:34 |
| *** mriedem_away is now known as mriedem | 13:35 | |
| johnthetubaguy | bauwser: folks are also asking for a way to get the current capacity, ways to "check" a new location for an instance violates and rules, etc | 13:35 |
| bauwser | johnthetubaguy: that's a separate concern | 13:35 |
| johnthetubaguy | bauwser: but these are all things that don't need claims, so its fine | 13:35 |
| bauwser | johnthetubaguy: my take is I would like to propose something like check_my_choice() | 13:35 |
| johnthetubaguy | bauwser: you can't "claim" a power usage level | 13:35 |
| *** ankit_ag has quit IRC | 13:35 | |
| *** rajesht has quit IRC | 13:35 | |
| bauwser | johnthetubaguy: that's what we call metrics | 13:36 |
| bauwser | johnthetubaguy: compared to resources | 13:36 |
| bauwser | johnthetubaguy: a metric is just used for helping to find a target, a resource is something you can claim about | 13:36 |
| bauwser | sorry to be bothering on that, but I followed jaypipes's 101 :) | 13:37 |
| bauwser | johnthetubaguy: and monitors track metrics, so there is no contract | 13:37 |
| *** artom__ has joined #openstack-nova | 13:37 | |
| johnthetubaguy | bauwser: we are talking about health metrics here right? | 13:41 |
| *** artom_ has quit IRC | 13:41 | |
| johnthetubaguy | I maybe missed something | 13:41 |
| *** artom__ has quit IRC | 13:41 | |
| *** ajayaa has quit IRC | 13:42 | |
| *** ajayaa has joined #openstack-nova | 13:43 | |
| *** rushiagr is now known as rushiagr_away | 13:44 | |
| *** atuvenie has joined #openstack-nova | 13:45 | |
| *** Hosam has joined #openstack-nova | 13:45 | |
| *** burt has joined #openstack-nova | 13:47 | |
| *** smccully has quit IRC | 13:48 | |
| *** snikitin has quit IRC | 13:48 | |
| *** smccully has joined #openstack-nova | 13:49 | |
| *** armax has joined #openstack-nova | 13:50 | |
| *** apmelton_away is now known as apmelton | 13:50 | |
| *** nikhil_k-afk is now known as nikhil_k | 13:51 | |
| *** johnthetubaguy is now known as zz_johnthetubagu | 13:52 | |
| *** armax has quit IRC | 13:52 | |
| *** armax has joined #openstack-nova | 13:53 | |
| *** mwagner_lap has joined #openstack-nova | 13:53 | |
| *** diegows has quit IRC | 13:54 | |
| *** mriedem has quit IRC | 13:56 | |
| figleaf | for metrics, a claim would simply be a no-op | 13:57 |
| *** dansmith is now known as superdan | 13:59 | |
| *** otter768 has joined #openstack-nova | 13:59 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/176844 | 14:00 |
| *** emagana has quit IRC | 14:00 | |
| *** mpaolino has quit IRC | 14:00 | |
| *** mriedem has joined #openstack-nova | 14:01 | |
| *** artom has joined #openstack-nova | 14:03 | |
| *** alex_xu has quit IRC | 14:03 | |
| *** heyongli has quit IRC | 14:03 | |
| *** yuntong has quit IRC | 14:03 | |
| *** eliqiao has quit IRC | 14:03 | |
| *** otter768 has quit IRC | 14:04 | |
| *** signed8bit has joined #openstack-nova | 14:04 | |
| *** takedakn has quit IRC | 14:05 | |
| *** BadCub_Hiding is now known as BadCub | 14:06 | |
| *** sks has quit IRC | 14:08 | |
| flip214 | ERROR: InvocationError: '/home/jenkins/workspace/gate-nova-specs-docs/.tox/venv/bin/python setup.py build_sphinx' | 14:08 |
| flip214 | That's what Gerrit says on a spec submit... | 14:09 |
| flip214 | is it my fault? The spec itself gives a warning, but "InvocationError" sounds like wrong arguments. | 14:09 |
| *** afazekas has quit IRC | 14:09 | |
| mriedem | flip214: local or in gerrit upstream? | 14:09 |
| flip214 | gerrit upstream | 14:09 |
| mriedem | link? | 14:10 |
| flip214 | http://logs.openstack.org/53/134153/10/check/gate-nova-specs-docs/0740bbb/console.html | 14:10 |
| mriedem | http://logs.openstack.org/53/134153/10/check/gate-nova-specs-docs/0740bbb/console.html#_2015-04-24_13_24_07_813 | 14:10 |
| mriedem | sphinx.errors.SphinxWarning: /home/jenkins/workspace/gate-nova-specs-docs/doc/source/specs/liberty/approved/drbd-driver.rst:179: WARNING: Definition list ends without a blank line; unexpected unindent. | 14:10 |
| mriedem | it's a problem with your spec format | 14:10 |
| flip214 | although that only says "warning"? | 14:10 |
| mriedem | we fail on warning | 14:10 |
| *** cfriesen has joined #openstack-nova | 14:10 | |
| mriedem | http://git.openstack.org/cgit/openstack/nova-specs/tree/setup.cfg#n20 | 14:11 |
| flip214 | well, the exact same content was allowed last time... https://review.openstack.org/#/c/134153/9..10/specs/liberty/approved/drbd-driver.rst,unified | 14:12 |
| flip214 | let's try again. | 14:12 |
| openstackgerrit | Ph. Marek proposed openstack/nova-specs: Connecting Nova to DRBD storage nodes directly. https://review.openstack.org/134153 | 14:12 |
| flip214 | sigh | 14:12 |
| mriedem | flip214: are you getting the same thing locally? | 14:12 |
| mriedem | tox -r -e docs | 14:12 |
| *** armax has quit IRC | 14:13 | |
| *** armax has joined #openstack-nova | 14:14 | |
| flip214 | rbash: tox: command not found | 14:14 |
| *** alex_xu has joined #openstack-nova | 14:14 | |
| *** heyongli has joined #openstack-nova | 14:14 | |
| mriedem | umm | 14:15 |
| mriedem | sudo pip install tox | 14:15 |
| flip214 | I'll just hope that Gerrit says "okay" this time... | 14:16 |
| *** eliqiao has joined #openstack-nova | 14:16 | |
| *** thangp has joined #openstack-nova | 14:16 | |
| *** yuntong has joined #openstack-nova | 14:17 | |
| *** thangp has quit IRC | 14:17 | |
| *** tonytan4ever has joined #openstack-nova | 14:17 | |
| *** thangp has joined #openstack-nova | 14:17 | |
| *** ajayaa has quit IRC | 14:19 | |
| *** zz_johnthetubagu is now known as johnthetubaguy | 14:20 | |
| flip214 | mriedem: thank you for clearing up the Warning/Error confusion! | 14:20 |
| openstackgerrit | Ph. Marek proposed openstack/nova-specs: Connecting Nova to DRBD storage nodes directly. https://review.openstack.org/134153 | 14:22 |
| bauwser | johnthetubaguy: sorry was diverted, by speaking of metrics, I just would like to say something that can be used in a filter or in a weighter without explicetely being necessary to be claimed | 14:22 |
| bauwser | johnthetubaguy: health metrics match that definition, I agree | 14:23 |
| bauwser | johnthetubaguy: but that's not all of that | 14:23 |
| *** mtanino has joined #openstack-nova | 14:27 | |
| *** oro has joined #openstack-nova | 14:29 | |
| *** dguitarbite has quit IRC | 14:30 | |
| openstackgerrit | Petrut Lucian proposed openstack/nova-specs: Hyper-V: Add storage QoS support https://review.openstack.org/175422 | 14:32 |
| *** Hosam has quit IRC | 14:32 | |
| *** zz_jgrimm is now known as jgrimm | 14:33 | |
| *** Hosam has joined #openstack-nova | 14:33 | |
| *** Nikolay_St has joined #openstack-nova | 14:38 | |
| *** sudipto has joined #openstack-nova | 14:38 | |
| *** jichen has quit IRC | 14:42 | |
| *** foexle has quit IRC | 14:42 | |
| *** foexle has joined #openstack-nova | 14:43 | |
| *** jwcroppe has joined #openstack-nova | 14:47 | |
| *** Nikolay_St has quit IRC | 14:47 | |
| *** Nikolay_St has joined #openstack-nova | 14:48 | |
| *** foexle has quit IRC | 14:49 | |
| *** Marga_ has joined #openstack-nova | 14:50 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: neutron: log hypervisor_macs before raising PortNotUsable https://review.openstack.org/175458 | 14:52 |
| *** matrohon has quit IRC | 14:53 | |
| *** obondarev has quit IRC | 14:53 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Log a warning if sec-groups and port id are provided on boot https://review.openstack.org/154068 | 14:53 |
| *** obondarev has joined #openstack-nova | 14:54 | |
| openstackgerrit | Julien Danjou proposed openstack/nova: simple_tenant_usage: declare method static https://review.openstack.org/165017 | 14:56 |
| openstackgerrit | Julien Danjou proposed openstack/nova: Leverage timeutils, drop strtime() https://review.openstack.org/164753 | 14:56 |
| *** nelsnelson has joined #openstack-nova | 14:57 | |
| *** nelsnelson has quit IRC | 14:58 | |
| *** zzzeek has joined #openstack-nova | 14:58 | |
| openstackgerrit | Petrut Lucian proposed openstack/nova-specs: Hyper-V: Add storage QoS support https://review.openstack.org/175422 | 14:59 |
| *** achanda has joined #openstack-nova | 15:03 | |
| *** Bsony_ has joined #openstack-nova | 15:04 | |
| *** browne has joined #openstack-nova | 15:05 | |
| *** Bsony has quit IRC | 15:06 | |
| *** jungleboyj has joined #openstack-nova | 15:06 | |
| johnthetubaguy | bauwser: sorry, I dropped to apply patches, and talk to folks about networking, anyways, sorry, you were talking to me? | 15:07 |
| *** davideagnello has quit IRC | 15:07 | |
| bauwser | johnthetubaguy: nah, just following up on our convo, but that's fine :) | 15:07 |
| bauwser | johnthetubaguy: we are all busy :) | 15:07 |
| *** achanda has quit IRC | 15:08 | |
| bauwser | johnthetubaguy: that said, I reinforce my interest in your devref change, if you plan to do such :) | 15:08 |
| mriedem | ndipanoff: have you heard of any bugs where attaching a volume (w/o specifying the device to use) fails because it picks a device already in use? | 15:08 |
| johnthetubaguy | mriedem: I have seen that with XenAPI FWIW, normally its BDM thats not in sync, due to races or otherwise | 15:09 |
| superdan | mriedem: if the device name is specified, it's easy to trip over that with libvirt | 15:09 |
| mriedem | device name isn't specified | 15:09 |
| mriedem | jungleboyj: ^ you have details? | 15:09 |
| johnthetubaguy | mriedem: do two and once, they probably race adding the BDM I guess? | 15:09 |
| johnthetubaguy | bauwser: I keep running out of hours in a day, but yeah, I need to start that | 15:10 |
| bauwser | johnthetubaguy: eh :) | 15:10 |
| jungleboyj | superdan: In this case I am trying to do 'nova volume-attach ... auto' . | 15:10 |
| *** sandywalsh has quit IRC | 15:10 | |
| johnthetubaguy | superdan: I am getting told jesse is getting angry about this on twitter: https://wiki.openstack.org/w/index.php?title=ReleaseNotes/Kilo&diff=prev&oldid=75612 | 15:11 |
| superdan | specifying auto? | 15:11 |
| johnthetubaguy | superdan: thats no reason to jump up and down, but do you have the lp bug id for that? | 15:11 |
| jungleboyj | superdan: It is picking /dev/vdb and when we get to libvirt it says the device is in use. | 15:11 |
| superdan | johnthetubaguy: jesse who? | 15:11 |
| johnthetubaguy | superdan: at blue host I think, let me find link | 15:11 |
| jungleboyj | I had sworn I had seen that as a known issue in the past. Could be wrong though. | 15:11 |
| superdan | johnthetubaguy: angry that it's broken or what? | 15:12 |
| johnthetubaguy | superdan: https://twitter.com/iamjkeating/status/591420824424316928 | 15:12 |
| johnthetubaguy | superdan: folks moaning about us shipping with a bug like that | 15:12 |
| mriedem | johnthetubaguy: it's not a new bug | 15:12 |
| superdan | johnthetubaguy: well, they should moan about every release since the feature was introduced :) | 15:12 |
| kashyap | mriedem: I can run a test here, if you have a bug pointer | 15:12 |
| mriedem | it's not a regression | 15:12 |
| kashyap | (Test w/ current git, ie) | 15:13 |
| superdan | right, the feature was the regression | 15:13 |
| superdan | johnthetubaguy: https://review.openstack.org/#/c/161444/ | 15:13 |
| mriedem | if only i were on twitter to correct | 15:13 |
| *** baoli has quit IRC | 15:13 | |
| superdan | mriedem: it's gone far beyond correctable state :) | 15:14 |
| johnthetubaguy | superdan: I just wonder about defaulting to not deleting them, and instead keep them powered off instead, but I am probably over thinking it | 15:14 |
| johnthetubaguy | superdan: yeah, agreed with the fixing it properly concern | 15:14 |
| superdan | johnthetubaguy: that breaks the api | 15:14 |
| superdan | johnthetubaguy: and then we need to clean them somehow, which we can't do without a new API or new guessing | 15:14 |
| johnthetubaguy | superdan: well it would just leave orphaned VMs I guess, but yeah, thats true… | 15:14 |
| bauwser | johnthetubaguy: superdan: mmm, I wonder if Nova could/should provide an API for such a rename | 15:14 |
| superdan | yeah, but that's like the worst thing ever, right? | 15:14 |
| bauwser | johnthetubaguy: superdan: because I often have questions about that | 15:15 |
| mriedem | kashyap: jungleboyj was the one asking me about it | 15:15 |
| superdan | because we leave customer data laying around on random hosts, disk fill quickly, etc | 15:15 |
| *** annashen has joined #openstack-nova | 15:15 | |
| johnthetubaguy | superdan: depends, worst thing ever is deleting customer data to some folks | 15:15 |
| superdan | the feature is *completely* broken right now | 15:15 |
| superdan | it just needs to be fixed | 15:15 |
| lascii | yeah, becomes a DoS vector | 15:15 |
| johnthetubaguy | superdan: agreed its borken | 15:15 |
| superdan | johnthetubaguy: heh, well, second worst then :D | 15:15 |
| johnthetubaguy | lascii: its an admin api I guess, but yes | 15:16 |
| superdan | johnthetubaguy: how about I update the relnotes to mention that this isn't a regression and has been the case since the feature was introduced? | 15:16 |
| jungleboyj | kashyap: I can put a bug together for it. Hadn't done that as I swore I saw one in the past, but I must be wrong. | 15:16 |
| superdan | that should help soothe the concerns that this is something that will suddenly start biting people? | 15:16 |
| mriedem | the PMs are all up in arms on the twitter :) | 15:16 |
| mriedem | shock and awe | 15:16 |
| mriedem | evacuate broken?! FML LOL :P | 15:17 |
| mriedem | that would be my response if i were on there | 15:17 |
| kashyap | jungleboyj: I'd suggest to hold off filing a bug unless you have a crystal clear reproducer. | 15:17 |
| *** Nikolay_St has quit IRC | 15:17 | |
| johnthetubaguy | superdan: I like that idea, can you link to a bug / spec for the fix too? | 15:17 |
| superdan | sure | 15:17 |
| jungleboyj | kashyap: Yeah, I was going to go try again on that system to get info for the bug. | 15:17 |
| * kashyap is not on twitter, seems like he's not missing out that much :-) | 15:17 | |
| *** Bsony_ has quit IRC | 15:17 | |
| mriedem | jungleboyj: this is libvirt? | 15:18 |
| jungleboyj | kashyap: You aren't missing much. I tried it for a while and just couldn't get into it. | 15:18 |
| jungleboyj | mriedem: Yes. | 15:18 |
| superdan | johnthetubaguy: https://wiki.openstack.org/wiki/ReleaseNotes/Kilo#Known_Issues_2 | 15:19 |
| *** Nikolay_St has joined #openstack-nova | 15:20 | |
| johnthetubaguy | superdan: sorry to be a pain the arse, thanks for the update on that | 15:20 |
| superdan | johnthetubaguy: np | 15:20 |
| *** annashen has quit IRC | 15:21 | |
| johnthetubaguy | superdan: I was tempted to just reply patches welcome, but I thought better of it | 15:21 |
| *** david-ly_ is now known as david-lyle | 15:22 | |
| superdan | johnthetubaguy: yeah, it's a twitter storm now, no amount of sanity will stop it :) | 15:22 |
| *** eliqiao1 has joined #openstack-nova | 15:23 | |
| mriedem | jungleboyj: you're not actually passing 'auto' as the device to the attach api are you? http://git.openstack.org/cgit/openstack/nova/tree/nova/compute/api.py#n3020 | 15:23 |
| superdan | just like me saying "have you hit this so far? No? then don't be concerned because you've been running this code for years now :)" | 15:23 |
| *** dpaterson has joined #openstack-nova | 15:24 | |
| openstackgerrit | Anthony PERARD proposed openstack/nova: libvirt: Use qemu as block driver for Xen instances https://review.openstack.org/177311 | 15:24 |
| johnthetubaguy | superdan: :) | 15:25 |
| dpaterson | Is there a way to get tenant quota usage via cli? | 15:25 |
| *** eliqiao has quit IRC | 15:25 | |
| *** lucasagomes_ has joined #openstack-nova | 15:26 | |
| mriedem | dpaterson: http://docs.openstack.org/cli-reference/content/novaclient_commands.html#novaclient_subcommand_usage | 15:27 |
| mriedem | jungleboyj: this is where the device name is picked based on the bdms on the instance | 15:28 |
| mriedem | http://git.openstack.org/cgit/openstack/nova/tree/nova/compute/utils.py#n94 | 15:28 |
| *** lucasagomes has quit IRC | 15:28 | |
| *** lucasagomes_ is now known as lucasagomes | 15:28 | |
| *** Nikolay_St has quit IRC | 15:30 | |
| dpaterson | mriedem: thanks, not quite what I was looking for, trying to find equivalent to cinder quota-usage, doesn't seem to be a cli call for that. | 15:31 |
| *** jlvillal has quit IRC | 15:31 | |
| dpaterson | in nova | 15:31 |
| mriedem | dpaterson: http://docs.openstack.org/cli-reference/content/novaclient_commands.html ? | 15:33 |
| mriedem | dpaterson: absolute-limits should give limits and usage | 15:34 |
| dpaterson | mriedem: that's it! tx | 15:37 |
| *** foexle has joined #openstack-nova | 15:37 | |
| *** pask81 has quit IRC | 15:39 | |
| *** andreykurilin__ has joined #openstack-nova | 15:40 | |
| *** ajayaa has joined #openstack-nova | 15:40 | |
| mriedem | jungleboyj: is this a multi-attach case you're dealing with? | 15:42 |
| mriedem | where there could be a race | 15:42 |
| *** Bsony has joined #openstack-nova | 15:43 | |
| openstackgerrit | dave-mcnally proposed openstack/nova: Add policy to cover snapshotting of volume backed instances https://review.openstack.org/176805 | 15:44 |
| jungleboyj | mriedem: No. And to clarify the confusion. I see the same behavior if I omit the 'auto' option on the CLI volume-attach command. | 15:45 |
| *** Marga_ has quit IRC | 15:45 | |
| *** lpetrut has quit IRC | 15:45 | |
| ndipanoff | mriedem, sorry was away | 15:46 |
| ndipanoff | yes I have | 15:46 |
| mriedem | jungleboyj: ah, | 15:46 |
| mriedem | def do_volume_attach(cs, args): | 15:46 |
| mriedem | """Attach a volume to a server.""" | 15:46 |
| mriedem | if args.device == 'auto': | 15:46 |
| mriedem | args.device = None | 15:46 |
| mriedem | that's the cli | 15:46 |
| mriedem | so it translates that for you | 15:46 |
| ndipanoff | jungleboyj, are you using libvirt?? | 15:47 |
| jungleboyj | ndipanoff: Yes. | 15:48 |
| jungleboyj | mriedem: Right. At one point in the past I think you had to specify auto, looks like that was fixed. :-) | 15:48 |
| ndipanoff | jungleboyj, did you report a bug? | 15:49 |
| *** ajayaa has quit IRC | 15:50 | |
| mriedem | ndipanoff: jungleboyj? https://bugs.launchpad.net/nova/+bug/1291835 | 15:50 |
| openstack | Launchpad bug 1291835 in OpenStack Compute (nova) "Repeated volume attach can cause u'message': u'The supplied device path (/dev/vdc) is in use.'" [Medium,Confirmed] | 15:50 |
| jungleboyj | ndipanoff: I am going to as it appears this isn't a known issue as I had thought. :-) | 15:50 |
| *** foexle has quit IRC | 15:50 | |
| *** exploreshaifali has quit IRC | 15:50 | |
| *** ajayaa has joined #openstack-nova | 15:50 | |
| ndipanoff | well I think I know what it is about - but would like to have a full bug report to be certain | 15:51 |
| *** sdake has joined #openstack-nova | 15:51 | |
| jungleboyj | ndipanoff: Ok, can do. Then at a minimum it is documented if others see it. | 15:51 |
| jungleboyj | I will get something pushed up today. | 15:52 |
| ndipanoff | mriedem, I don't think it's the same thing | 15:52 |
| ndipanoff | that's a race right | 15:52 |
| *** eliqiao1 has quit IRC | 15:52 | |
| ndipanoff | jungleboyj, you are not stress testing it like the bug above suggests right? | 15:53 |
| jungleboyj | ndipanoff: Shouldn't be the same as I am only doing one attach. Only person on the system. | 15:53 |
| ndipanoff | yep as I thought | 15:53 |
| ndipanoff | I seem to remember reading through the code when I was chasing a different bug and thinking that there is an inconsistency in tyhe way device names are defaulted | 15:54 |
| ndipanoff | but I also seem to remember deciding that it was not very likely to happen | 15:54 |
| *** Marga_ has joined #openstack-nova | 15:55 | |
| jungleboyj | ndipanoff: Interesting. I know I have seen this more than once. Will let you know when I get the bug in. | 15:55 |
| ndipanoff | so if you could give me a reproducer possibly with the select * from block_device_mapping where instance_uuid=<UUID> that would be awesome | 15:55 |
| ndipanoff | jungleboyj, many thanks! | 15:56 |
| mriedem | we could definitely use more debug logging in here http://git.openstack.org/cgit/openstack/nova/tree/nova/compute/utils.py#n125 | 15:56 |
| jungleboyj | ndipanoff: Thank you! | 15:56 |
| openstackgerrit | Christine Wang proposed openstack/nova: Return 404 if run volume ops with invalid volume https://review.openstack.org/165244 | 15:56 |
| jungleboyj | mriedem: Agreed. | 15:57 |
| ndipanoff | mriedem, agreed :) | 15:57 |
| beagles | superdan: regarding https://review.openstack.org/#/c/167287/ | 15:57 |
| mriedem | jungleboyj: ndipanoff: i could add something there | 15:58 |
| beagles | superdan, would you be opposed to separating the conversion to the NovaObject as a separate spec or patch? | 15:58 |
| jungleboyj | mriedem: I can try to do that as well as I am looking at this issue. | 15:58 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: add get_cpu_stats method to host https://review.openstack.org/177223 | 15:59 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: add get hardware info method to Host https://review.openstack.org/177222 | 15:59 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: add define_xml method to host https://review.openstack.org/174314 | 15:59 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: reuse already defined list of domains https://review.openstack.org/177328 | 15:59 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: add device_lookup_by_name to Host https://review.openstack.org/177329 | 15:59 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: add list_devices to Host https://review.openstack.org/177330 | 15:59 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: add compare_cpu to Host https://review.openstack.org/177331 | 15:59 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: add note to do note call libvirt from the driver https://review.openstack.org/177332 | 15:59 |
| mriedem | whatevs, | 15:59 |
| mriedem | i'm going to lunch now anyway, all this bdm talk has me hungry for a blt | 15:59 |
| *** vilobhmm1 has joined #openstack-nova | 15:59 | |
| *** tsekiyama has joined #openstack-nova | 15:59 | |
| jungleboyj | mriedem: :-) | 15:59 |
| *** mriedem is now known as mriedem_nom | 15:59 | |
| superdan | beagles: well, you're not proposing the conversion in that patch anyway are you? | 15:59 |
| *** otter768 has joined #openstack-nova | 16:00 | |
| *** andreykurilin__ has quit IRC | 16:00 | |
| beagles | superdan, I would like *not* to at this time - with the conversion to object as a follow up effort | 16:01 |
| *** vilobhmm11 has joined #openstack-nova | 16:01 | |
| ndipanoff | mriedem_nom, pro-tip: get some avocado in there too | 16:01 |
| *** jlanoux has quit IRC | 16:01 | |
| *** dpaterson has quit IRC | 16:01 | |
| superdan | beagles: well, here's the thing about that | 16:01 |
| ndipanoff | lol | 16:01 |
| *** baoli has joined #openstack-nova | 16:01 | |
| superdan | beagles: this is stored in our database and passed over the wire | 16:01 |
| superdan | beagles: so the fewer times we change things, the easier it is | 16:01 |
| superdan | beagles: so, if you don't change any of the format, then maybe it's okay to do that first | 16:02 |
| superdan | beagles: but I'm also not sure how that's less work | 16:02 |
| *** dpaterson has joined #openstack-nova | 16:03 | |
| *** Mike_D_laptop has joined #openstack-nova | 16:03 | |
| *** vilobhmm1 has quit IRC | 16:04 | |
| *** baoli_ has joined #openstack-nova | 16:04 | |
| *** eliqiao has joined #openstack-nova | 16:04 | |
| beagles | superdan, mmm... I was going to say we weren't modifying the format initially, but that's already wrong because we removed ovs_interfaceid. Which means we changed the contents of on disk data .. yeah I see your point. We are already looking at a migration type scenario. | 16:04 |
| *** otter768 has quit IRC | 16:04 | |
| superdan | beagles: the point of objects is to give you a facade to hide such changes from the upper layers | 16:05 |
| superdan | beagles: so doing that first gives you more flexibility, IMHO | 16:05 |
| *** baoli has quit IRC | 16:06 | |
| beagles | superdan, right | 16:06 |
| superdan | jaypipes: wudup | 16:06 |
| *** smccully has quit IRC | 16:08 | |
| sahid | superdan: thanks for commenting on https://review.openstack.org/#/c/168876/ ( i have repied to you, | 16:08 |
| sahid | if you still believe it makes no sense for you i will just abandon the change | 16:08 |
| sahid | (well no early response is expected here :) | 16:09 |
| *** arnaud___ has joined #openstack-nova | 16:09 | |
| superdan | sahid: thanks for replying to me, I have replied to you :) | 16:10 |
| ndipanoff | I would like to thank you both for replying to each other | 16:11 |
| *** ZZelle has quit IRC | 16:11 | |
| superdan | heh | 16:11 |
| *** smccully has joined #openstack-nova | 16:12 | |
| *** vilobhmm11 has quit IRC | 16:12 | |
| *** marun has joined #openstack-nova | 16:12 | |
| *** sandywalsh has joined #openstack-nova | 16:12 | |
| *** dprince has quit IRC | 16:13 | |
| sahid | ndipanoff: ahah | 16:14 |
| superdan | sahid: we could also ask maybe ndipanoff and jaypipes to comment on that | 16:14 |
| sahid | superdan: sure it would be appreciate to get more comment | 16:14 |
| sahid | ndipanoff, jaypipes when you have a moment if you can take a look at this please https://review.openstack.org/#/c/168876/ | 16:15 |
| sahid | it's about to remove the method get_all from servicegroup | 16:15 |
| *** jogo is now known as flashgordon | 16:18 | |
| *** hemnafk is now known as hemna | 16:19 | |
| *** tjones1 has joined #openstack-nova | 16:19 | |
| bauwser | sahid: superdan: can I just jump into the convo about the ServiceGroup.get_all() thing ? | 16:19 |
| superdan | bauwser: NO | 16:19 |
| sahid | ahah | 16:20 |
| bauwser | superdan: eh eh, week-end time, I have my firepit to burn so... | 16:20 |
| *** atuvenie_ has joined #openstack-nova | 16:20 | |
| *** atuvenie has quit IRC | 16:20 | |
| *** gtt116_ has quit IRC | 16:20 | |
| *** unicell has quit IRC | 16:20 | |
| *** gtt116__ has joined #openstack-nova | 16:20 | |
| *** browne has quit IRC | 16:20 | |
| superdan | heh | 16:21 |
| superdan | bauwser: of course you're welcome to comment there | 16:21 |
| bauwser | sahid: no really, I'm living a few comments now, but before leaving, just wanted to explain that there are some differences between Service.get_all() and ServiceGroup.get_all() | 16:21 |
| bauwser | while ServiceList.get_all() (sorry, made a typo) refers to all the existing referenced services created when the RPC manager is running, | 16:21 |
| *** tjones2 has joined #openstack-nova | 16:21 | |
| *** tjones1 has quit IRC | 16:22 | |
| openstackgerrit | sahid proposed openstack/nova: servicegroup: remove get_all method never used as public https://review.openstack.org/168876 | 16:22 |
| bauwser | ServiceGroup.get_all() should rather return the services from the driver PoV | 16:22 |
| sahid | just updated the commit message | 16:22 |
| bauwser | and that could be diferent | 16:22 |
| bauwser | sahid: ok commenting then | 16:22 |
| sahid | bauwser: please comment yes | 16:23 |
| jaypipes | sahid: sure, will do shortly. | 16:24 |
| sahid | bauwser: i have refered myself to this https://review.openstack.org/#/c/168876/3/nova/servicegroup/drivers/db.py,cm | 16:25 |
| sahid | jaypipes: cool thanks | 16:25 |
| *** oro has quit IRC | 16:26 | |
| bauwser | sahid: I used a Potato example to explain my thoughts | 16:28 |
| *** jwcroppe has quit IRC | 16:28 | |
| bauwser | sahid: basically, the 2 methods are just checking apples vs. bananas | 16:28 |
| openstackgerrit | Oleg Bondarev proposed openstack/nova: Log a warning if sec-groups and port id are provided on boot https://review.openstack.org/154068 | 16:28 |
| *** jwcroppe has joined #openstack-nova | 16:28 | |
| bauwser | sahid: the real problem is that we're underestimating ServiceGroup.get_all() just because the DB driver is the default one and we're lazy enough to not use it - just because we have ServiceList.get_all() | 16:29 |
| *** tjones2 has quit IRC | 16:29 | |
| *** tjones1 has joined #openstack-nova | 16:30 | |
| *** tjones1 has quit IRC | 16:30 | |
| *** tjones1 has joined #openstack-nova | 16:30 | |
| *** tjones1 has quit IRC | 16:30 | |
| bauwser | and to paraphrase someone else, that's not because it behaves like a duck that this is a duck | 16:31 |
| superdan | bauwser: were you making a dan quayle joke? | 16:31 |
| *** jcoufal has quit IRC | 16:31 | |
| bauwser | superdan: if so, I wish I would know him... | 16:31 |
| superdan | https://www.youtube.com/watch?v=Wdqbi66oNuI | 16:31 |
| superdan | our former vice president, ladies and gentlemen... | 16:31 |
| figleaf | superdan: I was reading the comment and thinking the same thing... | 16:31 |
| bauwser | eh | 16:31 |
| *** romainh has left #openstack-nova | 16:31 | |
| *** tjones1 has joined #openstack-nova | 16:32 | |
| *** tjones1 has left #openstack-nova | 16:32 | |
| superdan | figleaf: I'd never seen the video until just now whilst googling | 16:32 |
| *** patrickeast has joined #openstack-nova | 16:32 | |
| sahid | bauwser: ok if you prefer s/get_all/get_by_topic | 16:32 |
| superdan | only heard about it | 16:32 |
| figleaf | superdan: it was on every news cast at the time | 16:32 |
| bauwser | sahid: even that, ServiceList will just check DB | 16:32 |
| superdan | figleaf: right, I heard it on the radio at the time | 16:32 |
| bauwser | sahid: here, you need to have a call to something checking the real world | 16:33 |
| superdan | figleaf: being taken to school in the morning, IIRC :) | 16:33 |
| *** atuvenie_ has quit IRC | 16:33 | |
| bauwser | sahid: that's like having the choice between the blue pill and the red pill | 16:33 |
| *** jwcroppe has quit IRC | 16:33 | |
| figleaf | superdan: Quayle: HW Bush's assassination protection | 16:33 |
| superdan | heh | 16:33 |
| *** jcoufal has joined #openstack-nova | 16:33 | |
| bauwser | the blue pill will just give you the feeling that everything is OK | 16:34 |
| *** jcoufal has quit IRC | 16:36 | |
| *** jichen has joined #openstack-nova | 16:36 | |
| *** jistr has quit IRC | 16:36 | |
| *** lpetrut has joined #openstack-nova | 16:36 | |
| bauwser | so, speaking of firepit, I really have to fire my barbecue... | 16:36 |
| sahid | bauwser: i don't understand your explanation, if something is wonrg about to maintaint up to date service status with the dabatase perhaps we have to fill bug in launchpad | 16:36 |
| openstackgerrit | Bo Quan proposed openstack/nova: Correct the help text for the compute option. https://review.openstack.org/167805 | 16:36 |
| bauwser | sahid: no, because it's useful too | 16:37 |
| sahid | because i can imagine that 90% of operators use the db driver since it is by default | 16:37 |
| bauwser | sahid: the DB is the reference pointer to the real world | 16:37 |
| bauwser | sahid: if you remove the pointer, that's useless | 16:37 |
| sahid | also it's imortant to see where we need to retrieve all of the services | 16:38 |
| bauwser | sahid: so, everytime you're firing up a new service, its pointer is written in DB | 16:38 |
| bauwser | sahid: it's important to have a list of pointers, I agree | 16:38 |
| bauwser | and if you want to check the pointer/reference, that's enough | 16:39 |
| *** arnaud___ has quit IRC | 16:39 | |
| bauwser | but sometimes, you need more | 16:39 |
| sahid | bauwser: that is my point, where in nova we need to get all services ? | 16:39 |
| sahid | in a critical part ? | 16:40 |
| sahid | i don't think so | 16:40 |
| sahid | actually i'm pretty sure :) | 16:40 |
| openstackgerrit | Bo Quan proposed openstack/nova: Correct the help text for the compute option. https://review.openstack.org/167805 | 16:40 |
| sahid | that is why i believe all of that pain to maintain this method on each driver is not necessay | 16:41 |
| *** lucasagomes is now known as lucas-beer | 16:41 | |
| *** unicell has joined #openstack-nova | 16:42 | |
| openstackgerrit | Mitsuhiro Tanino proposed openstack/nova: Prevent unintentional detaching volume from instance https://review.openstack.org/166695 | 16:45 |
| *** nelsnelson has joined #openstack-nova | 16:45 | |
| *** atuvenie_ has joined #openstack-nova | 16:45 | |
| *** sahid has quit IRC | 16:45 | |
| *** nelsnelson has quit IRC | 16:46 | |
| *** vilobhmm1 has joined #openstack-nova | 16:47 | |
| *** vilobhmm1 has quit IRC | 16:47 | |
| *** lucas-beer has quit IRC | 16:47 | |
| *** vilobhmm1 has joined #openstack-nova | 16:47 | |
| *** vilobhmm1 has quit IRC | 16:47 | |
| *** baoli_ has quit IRC | 16:47 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: WIP Add a method to skip cells syncs on instance.save https://review.openstack.org/177356 | 16:48 |
| *** vilobhmm1 has joined #openstack-nova | 16:48 | |
| *** baoli has joined #openstack-nova | 16:48 | |
| lascii | bauwser: ^ | 16:48 |
| *** derekh has quit IRC | 16:48 | |
| *** tjones2 has joined #openstack-nova | 16:48 | |
| *** achanda has joined #openstack-nova | 16:52 | |
| *** vilobhmm1 has quit IRC | 16:53 | |
| *** sdake has quit IRC | 16:53 | |
| *** alanf-mc has joined #openstack-nova | 16:53 | |
| *** vilobhmm1 has joined #openstack-nova | 16:53 | |
| *** davideagnello has joined #openstack-nova | 16:54 | |
| *** bnemec is now known as beekneemech | 16:54 | |
| *** nelsnelson has joined #openstack-nova | 16:58 | |
| *** lucas-beer has joined #openstack-nova | 16:59 | |
| *** browne has joined #openstack-nova | 17:02 | |
| openstackgerrit | Baodong (Robert) Li proposed openstack/nova-specs: Support live migration with macvtap SR-IOV https://review.openstack.org/136077 | 17:02 |
| openstackgerrit | Brent Eagles proposed openstack/nova-specs: WIP: VIF plug script support for Nova https://review.openstack.org/162468 | 17:05 |
| *** exploreshaifali has joined #openstack-nova | 17:08 | |
| *** mriedem_nom is now known as mriedem_rmorse | 17:09 | |
| *** mriedem_rmorse is now known as mriedem | 17:09 | |
| cfriesen | is there a way to mark a blueprint as abandoned? | 17:09 |
| superdan | cfriesen: obsolete | 17:10 |
| cfriesen | superdan: thx | 17:10 |
| *** jwcroppe has joined #openstack-nova | 17:10 | |
| *** tjones2 has quit IRC | 17:12 | |
| *** sdake has joined #openstack-nova | 17:12 | |
| *** atuvenie_ has quit IRC | 17:13 | |
| *** moshele has joined #openstack-nova | 17:15 | |
| *** dprince has joined #openstack-nova | 17:18 | |
| *** tjones1 has joined #openstack-nova | 17:19 | |
| *** ajayaa has quit IRC | 17:20 | |
| bauwser | lascii: cool, I had no time to draw a patch for it, but sounds like you're going into the same direction | 17:20 |
| *** oro has joined #openstack-nova | 17:21 | |
| *** ociuhandu has joined #openstack-nova | 17:22 | |
| mriedem | andreykurilin: if you're around, i had a question in https://review.openstack.org/#/c/176817/ | 17:23 |
| *** Mike_D_laptop has quit IRC | 17:25 | |
| *** jwcroppe has quit IRC | 17:25 | |
| *** jwcroppe has joined #openstack-nova | 17:25 | |
| *** penick has joined #openstack-nova | 17:26 | |
| *** salv-orlando has quit IRC | 17:26 | |
| *** eglynn has quit IRC | 17:27 | |
| *** melwitt has joined #openstack-nova | 17:29 | |
| *** jwcroppe has quit IRC | 17:30 | |
| *** raildo has quit IRC | 17:32 | |
| *** raildo has joined #openstack-nova | 17:36 | |
| mriedem | dboik: sc68cal: is the ipv6 restriction only a neutron thing here? https://review.openstack.org/#/c/168701/ or does this also not work for nova-network? | 17:38 |
| mriedem | the bug description is basically written for the neutron case | 17:39 |
| mriedem | vishy: would you have any idea there? ^ | 17:39 |
| *** r-daneel has joined #openstack-nova | 17:42 | |
| *** vladikr has quit IRC | 17:44 | |
| *** beagles is now known as beagles_brb | 17:46 | |
| *** igordcard_ has quit IRC | 17:49 | |
| dboik | mriedem: Is IPv6 floating IP something that is supported in nova-network? I was not aware if it was. | 17:50 |
| *** vladikr has joined #openstack-nova | 17:50 | |
| *** harlowja_away is now known as harlowja | 17:51 | |
| dboik | mriedem: The bug was focused on Neutron because when dualstack devstack was enabled, a test was failing that ran Neutron but used the nova-network floating IP API | 17:51 |
| mriedem | dboik: i don't know if nova-network supports floating ipv6, hence the question :) | 17:52 |
| *** beagles_brb is now known as beagles | 17:52 | |
| *** atuvenie has joined #openstack-nova | 17:52 | |
| *** rushil_ has joined #openstack-nova | 17:53 | |
| *** rushil has quit IRC | 17:54 | |
| *** Marga_ has quit IRC | 17:56 | |
| dboik | mriedem: I was fairly certain it wasn't supported, and I haven't seen it documented anywhere. I'll try it out right now to be sure | 17:59 |
| *** atuvenie has quit IRC | 17:59 | |
| *** gabriel-bezerra has joined #openstack-nova | 18:01 | |
| *** otter768 has joined #openstack-nova | 18:01 | |
| *** igordcard has joined #openstack-nova | 18:01 | |
| *** igordcard_ has joined #openstack-nova | 18:01 | |
| gabriel-bezerra | Hi, where can I find the release schedule for Nova (and other openstack modules, if possible)? | 18:02 |
| gabriel-bezerra | When are, say, spec freeze and feature freeze going to be? | 18:03 |
| jroll | superdan: check it out, this is weird: https://gist.github.com/jimrollenhagen/b6b45aa43878cdc89d89 | 18:03 |
| jroll | see also the note at the bottom | 18:03 |
| gabriel-bezerra | I mean about Liberty. | 18:05 |
| mriedem | gabriel-bezerra: there is no liberty release schedule posted yet | 18:05 |
| mriedem | https://wiki.openstack.org/wiki/Releases | 18:05 |
| *** ZZelle has joined #openstack-nova | 18:05 | |
| mriedem | gabriel-bezerra: i would expect it to align somewhat simarly to the juno release schedule | 18:05 |
| *** otter768 has quit IRC | 18:05 | |
| mriedem | but with the spec and feature freeze being closer to what happened in kilo (for nova) | 18:06 |
| *** yassine has quit IRC | 18:06 | |
| *** sandywalsh has quit IRC | 18:06 | |
| *** busterswt has quit IRC | 18:07 | |
| superdan | jroll: this is neutron of course, right? | 18:08 |
| *** sandywalsh has joined #openstack-nova | 18:08 | |
| jroll | superdan: yes | 18:08 |
| openstackgerrit | melanie witt proposed openstack/nova: WIP: send object to cells instance_update_at_top https://review.openstack.org/176518 | 18:08 |
| superdan | jroll: in the exception handler for build, it's good on line 18 when we return from the cleanup, right? and it's only bad once we refresh instance after setting the task state and saving it, yes? | 18:10 |
| jroll | superdan: correct, it's false | 18:12 |
| openstackgerrit | John Garbutt proposed openstack/nova: devref: update around what nova scope https://review.openstack.org/177374 | 18:13 |
| superdan | jroll: so, you're using the async network allocator, isn't it plausible that you're racing with that coming back and having modified your instance object while you're doing other cleanup? | 18:13 |
| *** igordcard has quit IRC | 18:13 | |
| jroll | superdan: well, the allocation happened 10 minutes before (as seen in logs), and isn't running on the reschedule because network_allocated=True (which is the real problem here) | 18:14 |
| openstackgerrit | John Garbutt proposed openstack/nova: devref: update around what nova scope https://review.openstack.org/177374 | 18:14 |
| superdan | jroll: ten minutes? | 18:14 |
| *** igordcard has joined #openstack-nova | 18:15 | |
| *** johnthetubaguy is now known as zz_johnthetubagu | 18:15 | |
| jroll | superdan: 8, to be exact, long build times | 18:15 |
| superdan | yikes | 18:15 |
| jroll | $reasons. | 18:16 |
| jroll | pretty decent for bare metal though :P | 18:16 |
| *** busterswt has joined #openstack-nova | 18:16 | |
| jroll | and to be clear, we've brought the compute cluster down to one compute, so no races between them going on | 18:16 |
| superdan | jroll: so, if you look at the async handler, it's setting it on the instance.system_metadata, but then doing old-school instance_update, which doesn't actually save/refresh the object, which is the same object that the build thread still has | 18:16 |
| jroll | oooo | 18:17 |
| superdan | jroll: so maybe make that path do instance.save()? | 18:17 |
| jroll | ya | 18:17 |
| jroll | I didn't realize that was the old thing, I thought it was the new thing | 18:18 |
| *** annashen has joined #openstack-nova | 18:18 | |
| jroll | (and didn't realize it didn't actually save/refresh) | 18:18 |
| superdan | can't be | 18:18 |
| superdan | because _instance_update doesn't get the whole instance | 18:18 |
| superdan | now | 18:18 |
| jroll | mmm | 18:18 |
| superdan | lascii: wasn't there some reason we couldn't convert this async path, like maybe because $cells ? | 18:18 |
| gabriel-bezerra | mriedem: OK. Thank you. | 18:19 |
| *** tjones1 has quit IRC | 18:19 | |
| lascii | superdan: not cells, because the old run_instance path used it | 18:20 |
| superdan | ah | 18:20 |
| lascii | now that it's deprecated I think we can change it | 18:20 |
| superdan | lascii: well, it's using object syntax right above that line | 18:20 |
| superdan | lascii: so it seems like it'd be broken for that anyway | 18:20 |
| superdan | but yeah | 18:20 |
| lascii | heh, yeah | 18:20 |
| jroll | I'll give this a shot and see what happens | 18:21 |
| *** annashen has quit IRC | 18:23 | |
| *** exploreshaifali has quit IRC | 18:25 | |
| morgabra | ugggg | 18:25 |
| morgabra | yeah | 18:25 |
| *** Marga_ has joined #openstack-nova | 18:26 | |
| morgabra | jroll: I just scanned through, makes sense | 18:26 |
| *** salv-orlando has joined #openstack-nova | 18:27 | |
| *** arnaud___ has joined #openstack-nova | 18:27 | |
| *** tjones1 has joined #openstack-nova | 18:28 | |
| *** tjones1 has quit IRC | 18:30 | |
| *** tjones1 has joined #openstack-nova | 18:30 | |
| *** salv-orlando has quit IRC | 18:30 | |
| *** tjones1 has left #openstack-nova | 18:31 | |
| *** patrickeast has quit IRC | 18:31 | |
| *** patrickeast has joined #openstack-nova | 18:31 | |
| *** tjones1 has joined #openstack-nova | 18:31 | |
| *** moshele has quit IRC | 18:35 | |
| *** ijw has joined #openstack-nova | 18:40 | |
| *** jlvillal has joined #openstack-nova | 18:42 | |
| *** signed8bit is now known as signed8bit_ZZZzz | 18:43 | |
| dboik | mriedem: Did some digging with baoli - it looks like floating IPv6 is not supported. Nat rules are only applied to iptables, not ip6tables: https://github.com/openstack/nova/blob/095e9398ecf69ffdaeb929287d5f5f9a38257361/nova/network/linux_net.py#L814 | 18:45 |
| mriedem | dboik: cool, can you link that into the change in respones to my question? | 18:45 |
| dboik | mriedem: yep | 18:45 |
| *** Longgeek has quit IRC | 18:45 | |
| *** jwcroppe has joined #openstack-nova | 18:47 | |
| *** eglynn has joined #openstack-nova | 18:47 | |
| *** eglynn has quit IRC | 18:48 | |
| mriedem | dboik: -1 on https://review.openstack.org/#/c/168701/ because of the LOG.exception in the api extension code, we don't need that | 18:48 |
| mriedem | it's not an unexpected critical failure | 18:49 |
| dboik | mriedem: ack, will fix that | 18:49 |
| mriedem | then i'm +2 | 18:49 |
| mriedem | thanks | 18:49 |
| *** moshele has joined #openstack-nova | 18:50 | |
| *** moshele has quit IRC | 18:53 | |
| *** arnaud___ has quit IRC | 18:54 | |
| *** thangp has quit IRC | 18:59 | |
| *** ndipanoff has quit IRC | 19:01 | |
| *** jichen has quit IRC | 19:04 | |
| *** patrickeast has quit IRC | 19:04 | |
| *** ijw_ has joined #openstack-nova | 19:08 | |
| openstackgerrit | Andrew Boik proposed openstack/nova: Associate floating IPs with first v4 fixed IP if none specified https://review.openstack.org/168701 | 19:09 |
| *** ijw has quit IRC | 19:10 | |
| gabriel-bezerra | mriedem: Do you have any idea of when the spec freeze is going to be? Liberty-1? Liberty-2? | 19:12 |
| *** ZZelle_ has joined #openstack-nova | 19:12 | |
| mriedem | gabriel-bezerra: i don't know | 19:12 |
| *** ZZelle_ has quit IRC | 19:12 | |
| mriedem | i would hope l-2 sometime | 19:12 |
| gabriel-bezerra | mriedem: for nova, specifically. | 19:13 |
| mriedem | that's what we did in kilo | 19:13 |
| gabriel-bezerra | mriedem: Any plan to change it for liberty? Have you had any discussion or have anyone commented about it? | 19:14 |
| mriedem | gabriel-bezerra: this was in the meeting agenda yesterday https://wiki.openstack.org/wiki/Release_Cycle_Management/Liberty_Tracking | 19:14 |
| gabriel-bezerra | mriedem: OK. I'll check it. Thanks. | 19:15 |
| mriedem | gabriel-bezerra: i don't see anything in that which mentions how it impacts freezes though, if at all | 19:15 |
| mriedem | so i would assume l-2 is spec freeze | 19:15 |
| *** eharney has quit IRC | 19:15 | |
| mriedem | gabriel-bezerra: but that kind of stuff is generally not sorted out in detail until the summit | 19:16 |
| *** tjones2 has joined #openstack-nova | 19:16 | |
| mriedem | gabriel-bezerra: can you explain the need to know right now? | 19:16 |
| mriedem | generally the earlier your spec is up for review the better | 19:17 |
| mriedem | so don't wait until a week before spec freeze and expect your thing to be accepted | 19:17 |
| gabriel-bezerra | mriedem: Our team is working on a driver for Ironic and it needs some changes to Nova that might need specs to be approved. | 19:17 |
| gabriel-bezerra | mriedem: I'm collecting the spec freezes of Ironic, Nova and whatever other projects our driver might need (Neutron, maybe) | 19:18 |
| gabriel-bezerra | mriedem: so we can think of what we can do until October and write all the specs we need to keep working and releasing features for the driver. | 19:19 |
| *** igordcard has quit IRC | 19:20 | |
| jroll | gabriel-bezerra: curious, what nova changes would you need for an ironic driver? | 19:20 |
| gabriel-bezerra | jroll: we need some information from the flavor, for example, to configure the bare metal node as needed for the flavor the user selected. | 19:21 |
| jroll | gabriel-bezerra: ok, we have some things around capabilities that can handle that for you already, I believe, we can talk about it more in the ironic channel if you want as other people probably know more about it than me | 19:22 |
| gabriel-bezerra | jroll: the capabilities are not exactly what we need because they are used to match the node. In our case, we just need some setting from the flavor. | 19:23 |
| jroll | gabriel-bezerra: yeah, that's something we're working on :) | 19:23 |
| *** tsekiyama has quit IRC | 19:23 | |
| gabriel-bezerra | jroll: we could create a capability that could match anything, but that would be a kludge. | 19:23 |
| jroll | gabriel-bezerra: really, ironic wants to do this in a generic way, you should talk to ironic people about it | 19:24 |
| jroll | at least before you submit a nova spec | 19:24 |
| jroll | because I know other people want to do this and you could collaborate | 19:25 |
| gabriel-bezerra | jroll: Our team fixed it changing some code in Nova. The guys here created a new extra_spec namespace with information that should be just put in the call to ironic. | 19:26 |
| *** busterswt has quit IRC | 19:26 | |
| gabriel-bezerra | That's why I came to see nova's release schedule. | 19:26 |
| jroll | gabriel-bezerra: I'm just suggesting you work with others that have also worked on this problem, and collaborate, feel free to do it in a corner if you would like | 19:26 |
| mriedem | gabriel-bezerra: the nova release schedule has less to do with it | 19:27 |
| mriedem | gabriel-bezerra: liberty specs are open, so if you have specs you can submit them | 19:27 |
| mriedem | ironic people can then review the spec | 19:27 |
| gabriel-bezerra | jroll: I'm just commenting on what we have done. | 19:27 |
| *** signed8bit_ZZZzz is now known as signed8bit | 19:28 | |
| gabriel-bezerra | mriedem: but that should be a Nova spec, right? | 19:29 |
| gabriel-bezerra | We have another change in the information that is passed from nova to ironic. Would it be the same? | 19:30 |
| mriedem | gabriel-bezerra: https://wiki.openstack.org/wiki/Blueprints | 19:30 |
| mriedem | you need a blueprint and a change to review in the nova-specs repo | 19:30 |
| jroll | it's going to be both ironic and nova, right? I would say the ironic spec should land before the nova spec does, and the ironic spec should have some notes or a reference to how the nova side works | 19:30 |
| jroll | that's how we've done this in the past, anyhow | 19:30 |
| mriedem | gabriel-bezerra: http://lists.openstack.org/pipermail/openstack-dev/2015-March/058107.html | 19:30 |
| gabriel-bezerra | jroll: even if it only comprehends changes in Nova's codebase? | 19:31 |
| jroll | gabriel-bezerra: so there won't be changes in ironic associated with this? | 19:32 |
| *** tjones1 has quit IRC | 19:33 | |
| gabriel-bezerra | jroll: regarding the changes we need in nova we would piggyback on things that are already implemented in ironic, just add information to calls that are already made. | 19:33 |
| *** tsekiyama has joined #openstack-nova | 19:33 | |
| gabriel-bezerra | jroll: we would need a spec for the initial implementation of the driver anyway though | 19:34 |
| jroll | gabriel-bezerra: 1) any new driver in ironic requires a spec; 2) "created a new extra_spec namespace... put in the call to ironic" would change the api and probably require a spec as well | 19:34 |
| jroll | yeah | 19:34 |
| *** jwcroppe has quit IRC | 19:34 | |
| gabriel-bezerra | jroll: So you mean we should land the spec for the driver and then mention it in the specs for the changes we need in nova? | 19:35 |
| figleaf | "just add information" is a bit more complicated than it sounds | 19:35 |
| jroll | gabriel-bezerra: what I mean is, we shouldn't approve the nova spec until the ironic spec is approved, since the interactions may change during spec review | 19:36 |
| gabriel-bezerra | figleaf: the name of the instance, for example. | 19:36 |
| gabriel-bezerra | figleaf: and some data from the flavor. | 19:36 |
| gabriel-bezerra | jroll: I see. | 19:37 |
| gabriel-bezerra | jroll: the spec freeze includes spec approval freeze, right? not just spec submission freeze. | 19:37 |
| *** igordcar1_ has joined #openstack-nova | 19:37 | |
| figleaf | gabriel-bezerra: I understand that it's small things, but it's still changing an api and needs to be versioned | 19:37 |
| jroll | gabriel-bezerra: I don't know anything about the freezes for liberty yet because I think freezes are a joke | 19:38 |
| jroll | sorry | 19:38 |
| *** sudipto has quit IRC | 19:39 | |
| figleaf | jroll: :) | 19:39 |
| jroll | :) | 19:41 |
| *** jwcroppe has joined #openstack-nova | 19:41 | |
| *** igordcar1_ has quit IRC | 19:41 | |
| *** tjones1 has joined #openstack-nova | 19:42 | |
| *** jwcroppe has quit IRC | 19:42 | |
| gabriel-bezerra | figleaf: would it be troublesome to do that even if we are just adding and not removing any data from a call nova does to Ironic? | 19:42 |
| figleaf | gabriel-bezerra: yes | 19:43 |
| *** igordcar1_ has joined #openstack-nova | 19:43 | |
| jroll | gabriel-bezerra: I would say post your ironic spec ASAP and we can figure out where to go from there, because it's unclear to me exactly what you're doing | 19:43 |
| *** igordcard_ has quit IRC | 19:44 | |
| figleaf | gabriel-bezerra: removing data would be really, really bad, but adding data can still trip things up | 19:44 |
| *** igordcard_ has joined #openstack-nova | 19:44 | |
| figleaf | gabriel-bezerra: as jroll said, we'd need a spec to understand what the implications would be | 19:45 |
| *** igordcard has joined #openstack-nova | 19:45 | |
| *** igordcard_ has quit IRC | 19:45 | |
| *** igordcard_ has joined #openstack-nova | 19:45 | |
| gabriel-bezerra | figleaf, jroll, mriedem: OK. I see. I'll work a plan out here with the team for these specs and the release. Thank you for your help and advices. | 19:47 |
| openstackgerrit | Joel Coffman proposed openstack/nova: libvirt: Clean up unit tests for _hard_reboot https://review.openstack.org/177398 | 19:47 |
| *** igordcard has quit IRC | 19:47 | |
| jroll | gabriel-bezerra: you're welcome | 19:47 |
| *** eharney has joined #openstack-nova | 19:48 | |
| *** arnaud___ has joined #openstack-nova | 19:50 | |
| *** jwcroppe has joined #openstack-nova | 19:51 | |
| *** Bsony has quit IRC | 19:54 | |
| *** nelsnelson has quit IRC | 19:54 | |
| *** arnaud___ has quit IRC | 19:54 | |
| *** jwcroppe has quit IRC | 19:55 | |
| *** jwcroppe has joined #openstack-nova | 19:57 | |
| *** jungleboyj has quit IRC | 19:57 | |
| *** jwcroppe has quit IRC | 19:57 | |
| *** jwcroppe has joined #openstack-nova | 19:58 | |
| *** patrickeast has joined #openstack-nova | 19:59 | |
| *** jwcroppe_ has joined #openstack-nova | 20:00 | |
| *** tjones2 has quit IRC | 20:01 | |
| *** jwcroppe has quit IRC | 20:02 | |
| *** otter768 has joined #openstack-nova | 20:02 | |
| *** ociuhandu has quit IRC | 20:02 | |
| *** yassine has joined #openstack-nova | 20:03 | |
| *** moshele has joined #openstack-nova | 20:05 | |
| *** otter768 has quit IRC | 20:06 | |
| *** jungleboyj has joined #openstack-nova | 20:08 | |
| *** thangp has joined #openstack-nova | 20:11 | |
| cfriesen | can someone double-check whether I'm reading this right? virt.hardware._numa_fit_instance_cell() will return None if instance_cell.memory > host_cell.memory. This seems like it could give false negatives in the case where the host has very little free 4K pages available but lots of hugepages available, and the instance is using hugepages. | 20:12 |
| cfriesen | istance_cell.memory could be allocated using hugepages, but I think host_cell.memory only reflects the 4K pages | 20:13 |
| *** jgrimm is now known as zz_jgrimm | 20:14 | |
| *** vladikr has quit IRC | 20:16 | |
| *** Bsony has joined #openstack-nova | 20:17 | |
| *** dave-mccowan has quit IRC | 20:18 | |
| openstackgerrit | Joel Coffman proposed openstack/nova: libvirt: Remove unnecessary JSON conversions https://review.openstack.org/177437 | 20:19 |
| *** Bsony has quit IRC | 20:23 | |
| *** Hosam has quit IRC | 20:25 | |
| *** Hosam has joined #openstack-nova | 20:26 | |
| *** Bsony has joined #openstack-nova | 20:27 | |
| *** busterswt has joined #openstack-nova | 20:27 | |
| *** isd has joined #openstack-nova | 20:29 | |
| *** Hosam has quit IRC | 20:31 | |
| *** penick has quit IRC | 20:31 | |
| *** Bsony has quit IRC | 20:32 | |
| *** ivasev has quit IRC | 20:34 | |
| mriedem | lascii: do we have a cells bug tracking this? http://goo.gl/RRUvI9 | 20:40 |
| mriedem | is that the instance.save thing? | 20:40 |
| mriedem | it's not 100% failure, but it's an ugly trace | 20:40 |
| lascii | there's no bug for that yet that I know of | 20:43 |
| mriedem | k, i'll open one | 20:43 |
| lascii | could be the instance.save thing, but I'd need to dig further | 20:43 |
| *** oro has quit IRC | 20:45 | |
| *** penick has joined #openstack-nova | 20:45 | |
| mriedem | v | 20:46 |
| mriedem | https://bugs.launchpad.net/nova/+bug/1448302 | 20:46 |
| openstack | Launchpad bug 1448302 in OpenStack Compute (nova) "cells: intermittent KeyError when deleting instance metadata" [Low,Confirmed] | 20:46 |
| lascii | thanks | 20:46 |
| mriedem | i'm seeing a cells job failure on a patch of mine so digging into errors | 20:46 |
| *** thangp has quit IRC | 20:47 | |
| lascii | awesome | 20:47 |
| *** yassine has quit IRC | 20:47 | |
| lascii | I'm testing a fix for the instance.save race, would like to solve that soon to at least eliminate it from contention | 20:48 |
| *** rushil_ has quit IRC | 20:51 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: n-net: turn down log level when vif isn't found in deallocate_fixed_ip https://review.openstack.org/177450 | 20:53 |
| mriedem | ^ oldy but goody | 20:54 |
| mriedem | this isn't just the cells job, but it's spiking lately http://goo.gl/PsZHAu | 20:56 |
| mriedem | http://logs.openstack.org/58/175458/3/check/check-tempest-dsvm-cells/ce91d7e/logs/screen-n-cpu.txt.gz?level=TRACE#_2015-04-24_15_16_26_866 | 20:56 |
| mriedem | regression in the compute service code since yesterday? | 20:57 |
| mriedem | https://github.com/openstack/nova/commit/f0f158af375efcc91e7c75859fceeec0111f424b | 20:57 |
| *** armax has quit IRC | 20:57 | |
| *** patrickeast has quit IRC | 20:58 | |
| mtreinish | mriedem: superdan pushed a devstack change to call kill twice to ensure things really die | 20:58 |
| superdan | hell yes I did | 20:58 |
| mtreinish | there was a critical bug somwhere I think | 20:59 |
| *** jungleboyj has quit IRC | 20:59 | |
| mriedem | https://github.com/openstack/nova/commit/02795069d069c4fd90a71d6187ba6de45689e68b | 21:00 |
| *** moshele has quit IRC | 21:00 | |
| *** htruta has quit IRC | 21:01 | |
| sc68cal | mriedem: thanks for taking a look at 168701 | 21:01 |
| *** busterswt has quit IRC | 21:01 | |
| *** annashen has joined #openstack-nova | 21:02 | |
| mriedem | sure | 21:02 |
| jroll | superdan: I think that instance.save() did it | 21:03 |
| jroll | s/I/we/ I shouldn't take all this credit | 21:03 |
| superdan | jroll: \o/ | 21:03 |
| *** dpaterson has quit IRC | 21:04 | |
| *** sdake_ has joined #openstack-nova | 21:04 | |
| superdan | jroll: gonna patch it or want me to? | 21:04 |
| jroll | superdan: we'll put a patch up at some point, do you want a decent bug report too? | 21:04 |
| jroll | oh, either way | 21:04 |
| jroll | if I do it, that's one less core we need | 21:04 |
| superdan | jroll: doubt it'll have much trouble landing | 21:05 |
| jroll | yeah | 21:05 |
| mriedem | jroll: if you want to backport it, you need a bug report | 21:05 |
| jroll | either way, we can get it up today | 21:05 |
| *** raildo has quit IRC | 21:05 | |
| jroll | mriedem: right, so I don't personally care if it's backported, that's up to y'all | 21:05 |
| mriedem | s/need/should have/ | 21:05 |
| superdan | jroll: that's cool, I'll leave you the glory if you want it :) | 21:05 |
| jroll | lol, I'll leave it to JoshNang, he did a lot of the work on this too :P | 21:05 |
| *** andreykurilin__ has joined #openstack-nova | 21:06 | |
| mriedem | superdan: mtreinish: what was the devstack change in the last day that i can't find | 21:06 |
| mriedem | i only see this https://github.com/openstack-dev/devstack/commit/6cdb2e0f1a2c2429587f1e9187344cb26eb31812 | 21:07 |
| superdan | sdague had to recreate it | 21:07 |
| superdan | because of a weird depends-on thing | 21:07 |
| *** salv-orlando has joined #openstack-nova | 21:07 | |
| *** sdake has quit IRC | 21:08 | |
| superdan | https://review.openstack.org/#/c/176932/ | 21:08 |
| superdan | we only did it on stable/kilo because it's a workarond | 21:08 |
| superdan | the real fix (well not really) is an oslo sync | 21:08 |
| superdan | which is this: https://review.openstack.org/#/c/176777/ | 21:09 |
| superdan | which you voted on | 21:09 |
| superdan | because I told you to :) | 21:09 |
| mriedem | yeah i knew about that | 21:09 |
| mriedem | so trying to figure out wtf http://goo.gl/PsZHAu | 21:09 |
| mriedem | spike there | 21:09 |
| mriedem | b/c this cells run failed http://logs.openstack.org/58/175458/3/check/check-tempest-dsvm-cells/ce91d7e/console.html#_2015-04-24_15_38_45_075 | 21:10 |
| mriedem | i'm assuming b/c the service was down | 21:10 |
| mriedem | ah http://logs.openstack.org/58/175458/3/check/check-tempest-dsvm-cells/ce91d7e/logs/screen-n-api.txt.gz?level=TRACE#_2015-04-24_15_21_05_563 | 21:10 |
| mriedem | i rarely check n-api for traces | 21:10 |
| *** tonytan4ever has quit IRC | 21:10 | |
| mriedem | lascii: ^ | 21:11 |
| mriedem | that new? | 21:11 |
| *** artom has quit IRC | 21:11 | |
| lascii | yeah, haven't seen that before | 21:11 |
| *** salv-orl_ has joined #openstack-nova | 21:13 | |
| mriedem | k, it doesn't hit a lot | 21:13 |
| mriedem | but 90% failure when it does | 21:13 |
| lascii | that shouldn't be cells related though | 21:13 |
| mriedem | i'll get a bug open | 21:13 |
| mriedem | it is | 21:13 |
| mriedem | only hits on the cells job | 21:14 |
| mriedem | http://goo.gl/4mTUIj | 21:14 |
| mriedem | File "/opt/stack/new/nova/nova/compute/cells_api.py", line 226, in _handle_cell_delete | 21:14 |
| lascii | weird. that method just casts to cells and then passes back to compute/api delete | 21:14 |
| *** salv-orlando has quit IRC | 21:14 | |
| lascii | well, _local_delete | 21:14 |
| *** dboik_ has joined #openstack-nova | 21:15 | |
| mriedem | well, that's also in the same job that had http://logs.openstack.org/58/175458/3/check/check-tempest-dsvm-cells/ce91d7e/logs/screen-n-cpu.txt.gz?level=TRACE#_2015-04-24_15_16_26_866 | 21:15 |
| mriedem | https://bugs.launchpad.net/nova/+bug/1448316 | 21:17 |
| openstack | Launchpad bug 1448316 in OpenStack Compute (nova) "cells: Object action destroy failed because: host changed" [Medium,Confirmed] | 21:17 |
| *** dboik has quit IRC | 21:18 | |
| *** sdake has joined #openstack-nova | 21:19 | |
| *** dboik_ has quit IRC | 21:19 | |
| *** salv-orl_ has quit IRC | 21:20 | |
| *** oro has joined #openstack-nova | 21:21 | |
| *** sdake_ has quit IRC | 21:23 | |
| *** Marga_ has quit IRC | 21:23 | |
| *** figleaf is now known as edleafe | 21:25 | |
| *** jungleboyj has joined #openstack-nova | 21:37 | |
| *** arnaud___ has joined #openstack-nova | 21:39 | |
| *** signed8bit has quit IRC | 21:40 | |
| *** joefides has quit IRC | 21:40 | |
| *** isd has quit IRC | 21:42 | |
| *** arnaud___ has quit IRC | 21:43 | |
| mriedem | lascii: on those big KeyError traces, i think they are probably due to negative tests in here, e.g. http://git.openstack.org/cgit/openstack/tempest/tree/tempest/api/compute/test_authorization.py#n394 | 21:43 |
| *** mriedem has quit IRC | 21:47 | |
| bauwser | oh mriedem left before I had time to thank him for the cells failures... | 21:49 |
| *** vilobhmm11 has joined #openstack-nova | 21:51 | |
| *** smccully has quit IRC | 21:52 | |
| *** Hosam has joined #openstack-nova | 21:52 | |
| *** vilobhmm12 has joined #openstack-nova | 21:53 | |
| *** vilobhmm1 has quit IRC | 21:53 | |
| *** dboik has joined #openstack-nova | 21:54 | |
| *** vilobhmm11 has quit IRC | 21:56 | |
| *** burt has quit IRC | 21:57 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: WIP Add a method to skip cells syncs on instance.save https://review.openstack.org/177356 | 21:58 |
| *** dave-mccowan has joined #openstack-nova | 21:59 | |
| *** baoli has quit IRC | 22:00 | |
| *** neelashah has quit IRC | 22:01 | |
| *** lpetrut has quit IRC | 22:02 | |
| openstackgerrit | Josh Gachnang proposed openstack/nova: Reschedules sometimes do not allocate networks https://review.openstack.org/177470 | 22:02 |
| JoshNang | jroll: ^ | 22:02 |
| jroll | superdan: ^^ | 22:02 |
| *** otter768 has joined #openstack-nova | 22:03 | |
| superdan | foobar: ^^^ | 22:03 |
| jroll | :P | 22:03 |
| *** mwagner_lap has quit IRC | 22:03 | |
| bauwser | lascii: had no time yet to run a review for your patch | 22:03 |
| lascii | bauwser: np, it still needs some work | 22:04 |
| bauwser | lascii: I also saw melwitt providing a patch | 22:04 |
| bauwser | on top of it | 22:04 |
| bauwser | lascii: I leave you work on that bug while I'm trying to rework on the ReqSpec object ;) | 22:04 |
| lascii | bauwser: heh, np | 22:05 |
| melwitt | bauwser, lascii: yes, I had a patch going that was stuck to the save cycle problem, so I'm testing out with lascii patch | 22:05 |
| lascii | yeah, she's doing a lot of object work in cells with that patch. | 22:05 |
| lascii | oh, hi :) | 22:05 |
| melwitt | :) | 22:06 |
| bauwser | saw mriedem email, responded to it, but that furiously looks like "do_sth_with_instance(); do_sth_else_with_that_instance()" | 22:06 |
| bauwser | in the Tempest classes | 22:06 |
| lascii | melwitt: my patch worked with local conductor, but remote conductor needs a bit more work. | 22:06 |
| bauwser | so, there are high chances that a related instance.save() on the API is just beaten by the other one | 22:06 |
| *** otter768 has quit IRC | 22:07 | |
| bauwser | lascii: yeh, that's what I was working on | 22:07 |
| bauwser | lascii: because when serializing the object, it just misses any non field | 22:07 |
| lascii | yep, that's what I realized | 22:08 |
| bauwser | lascii: so I was just wondering how to serialize that attribute correctly | 22:08 |
| melwitt | lascii, bauwser: oh, interesting. that's probably what I'm seeing then, I am getting a lack of update to vm_state and task_state at the top with the original version | 22:08 |
| lascii | so I added it to the fields as a non_column field which should work, but tests are broken and I need to track that down | 22:08 |
| lascii | bauwser: ^ | 22:09 |
| bauwser | melwitt: lascii: don't miss the fact that when using the conductor, it's serialied/unserialized | 22:09 |
| bauwser | lascii: yeah, or an extra field should work too | 22:09 |
| melwitt | bauwser: ah, yeah | 22:09 |
| *** patrickeast has joined #openstack-nova | 22:09 | |
| bauwser | lascii: glad to see you're now facing the same problem I had | 22:10 |
| lascii | bauwser: extra fields don't appear to be serialized | 22:10 |
| bauwser | lascii: my bad we didn't had time to sync on our work | 22:10 |
| bauwser | lascii: orly? | 22:10 |
| lascii | no worries, I've been swamped recently | 22:10 |
| lascii | bauwser: not completely sure, but I don't think so | 22:10 |
| bauwser | lascii: yeah that's just bad that you took some time for finding that while I was already knowing the issue :( | 22:11 |
| lascii | fields plus adding it to _INSTANCE_OPTIONAL_NON_COLUMN_FIELDS should work I think | 22:11 |
| bauwser | lascii: hell yeah https://github.com/openstack/nova/blob/master/nova/objects/base.py#L496 | 22:11 |
| *** sdake_ has joined #openstack-nova | 22:11 | |
| *** Marga_ has joined #openstack-nova | 22:13 | |
| lascii | gotta run, but will start tracking down some bugs monday | 22:13 |
| *** lascii is now known as alaski | 22:13 | |
| bauwser | alaski: yeah was confounding with https://github.com/openstack/nova/blob/master/nova/objects/base.py#L615 | 22:13 |
| bauwser | alaski: ack, see ya, have a good WE | 22:14 |
| melwitt | seeya alaski | 22:14 |
| alaski | bauwser: yeah. I tried extra fields first too, then checked the serializer | 22:14 |
| alaski | bye | 22:15 |
| *** annashen has quit IRC | 22:15 | |
| *** r-daneel has quit IRC | 22:15 | |
| *** sdake has quit IRC | 22:15 | |
| *** oro has quit IRC | 22:18 | |
| *** thangp has joined #openstack-nova | 22:19 | |
| *** dave-mccowan has quit IRC | 22:19 | |
| *** foexle has joined #openstack-nova | 22:22 | |
| *** sdake has joined #openstack-nova | 22:23 | |
| *** yamahata has joined #openstack-nova | 22:25 | |
| *** sdake_ has quit IRC | 22:27 | |
| *** dprince has quit IRC | 22:28 | |
| mikal | . | 22:40 |
| *** harlowja_ has joined #openstack-nova | 22:41 | |
| *** erlon has quit IRC | 22:41 | |
| *** joefides has joined #openstack-nova | 22:41 | |
| *** harlowja has quit IRC | 22:41 | |
| cburgess | Do we store the scheduler hints from launch anywhere in the DB? I know we didn't use to but I can't recall if we added that or if thats all future new scheduler stuff. | 22:42 |
| *** Hosam has quit IRC | 22:42 | |
| *** jungleboyj has quit IRC | 22:44 | |
| superdan | no | 22:45 |
| superdan | part of the request spec spec I think | 22:45 |
| cburgess | k... I just couldn't remember if that was done in K or is part of the future new scheduler. stuff. | 22:45 |
| cburgess | Thanks. | 22:45 |
| *** angdraug has joined #openstack-nova | 22:47 | |
| *** marun has quit IRC | 22:47 | |
| *** panbalag has joined #openstack-nova | 22:49 | |
| *** xyang1 has quit IRC | 22:50 | |
| panbalag | Hi..I'm trying nova with SSL enabled and seeing the failure "error:14094410:SSL routines:SSL3_READ_BYTES:sslv3 alert handshake failure" while trying any nova command. Has anyone seen this issue before? | 22:50 |
| *** sdake_ has joined #openstack-nova | 22:53 | |
| *** achanda has quit IRC | 22:53 | |
| *** sdake__ has joined #openstack-nova | 22:55 | |
| melwitt | cburgess: there is this spec about persisting the request spec https://review.openstack.org/#/c/176078/ | 22:56 |
| *** sdake has quit IRC | 22:56 | |
| *** apmelton is now known as apmelton_away | 22:58 | |
| *** sdake_ has quit IRC | 22:59 | |
| *** igordcard has joined #openstack-nova | 23:02 | |
| *** hemna is now known as hemnafk | 23:09 | |
| cburgess | cool thanks | 23:12 |
| *** ganso_ has quit IRC | 23:17 | |
| *** arnaud___ has joined #openstack-nova | 23:28 | |
| superdan | panbalag: see topic | 23:32 |
| *** arnaud___ has quit IRC | 23:32 | |
| panbalag | superdan, I don't get the IRC name - is it #openstack-master? or just #openstack | 23:35 |
| superdan | #openstack | 23:35 |
| panbalag | superdan, Thanks. I'll post it there | 23:36 |
| *** takedakn has joined #openstack-nova | 23:43 | |
| *** annashen has joined #openstack-nova | 23:44 | |
| *** takedakn has quit IRC | 23:46 | |
| *** annashen has quit IRC | 23:48 | |
| *** thangp has quit IRC | 23:50 | |
| *** isd has joined #openstack-nova | 23:50 | |
| *** sdake__ has quit IRC | 23:52 | |
| *** achanda has joined #openstack-nova | 23:53 | |
| *** achanda has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!