| *** achanda has quit IRC | 00:00 | |
| *** achanda_ has quit IRC | 00:01 | |
| openstackgerrit | lalit dagre proposed openstack/nova: Flavor object which support both DB with active migration https://review.openstack.org/213041 | 00:02 |
|---|---|---|
| *** dims has joined #openstack-nova | 00:03 | |
| *** oomichi has joined #openstack-nova | 00:03 | |
| *** bnemec has quit IRC | 00:03 | |
| *** salv-orlando has quit IRC | 00:04 | |
| *** dims has quit IRC | 00:05 | |
| *** dims has joined #openstack-nova | 00:05 | |
| *** dave-mccowan has joined #openstack-nova | 00:06 | |
| *** takedakn has quit IRC | 00:07 | |
| *** takedakn has joined #openstack-nova | 00:08 | |
| *** vilobhmm1 has joined #openstack-nova | 00:09 | |
| *** tsekiyam_ has joined #openstack-nova | 00:09 | |
| *** dims has quit IRC | 00:10 | |
| *** thorst1 has quit IRC | 00:10 | |
| *** mtanino has quit IRC | 00:12 | |
| *** tsekiyama has quit IRC | 00:13 | |
| *** tsekiyam_ has quit IRC | 00:13 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Fix cinder volume stuck in swap_volume https://review.openstack.org/207385 | 00:13 |
| *** dims has joined #openstack-nova | 00:17 | |
| *** beagles has quit IRC | 00:22 | |
| *** takedakn has quit IRC | 00:22 | |
| *** meghal1 has quit IRC | 00:24 | |
| *** ubuntu has joined #openstack-nova | 00:25 | |
| *** ubuntu is now known as Guest172 | 00:25 | |
| *** sachin has quit IRC | 00:27 | |
| openstackgerrit | lalit dagre proposed openstack/nova: Flavor object which support both DB with active migration https://review.openstack.org/213041 | 00:29 |
| *** yuntongjin has joined #openstack-nova | 00:30 | |
| *** Guest172 has quit IRC | 00:31 | |
| *** asettle has joined #openstack-nova | 00:32 | |
| *** asettle has quit IRC | 00:33 | |
| *** asettle has joined #openstack-nova | 00:33 | |
| *** penick has quit IRC | 00:36 | |
| *** markvoelker has joined #openstack-nova | 00:39 | |
| *** tjones has left #openstack-nova | 00:46 | |
| *** browne has quit IRC | 00:54 | |
| *** vilobhmm1 has quit IRC | 00:56 | |
| *** yuntongjin1 has joined #openstack-nova | 00:56 | |
| *** EinstCrazy has joined #openstack-nova | 00:56 | |
| *** zhangjn has joined #openstack-nova | 00:57 | |
| *** zhenguo has joined #openstack-nova | 00:57 | |
| *** yuntongjin has quit IRC | 00:59 | |
| *** summer has joined #openstack-nova | 01:02 | |
| *** dannywilson has quit IRC | 01:03 | |
| *** winston-d_zZZ is now known as winston-d | 01:11 | |
| *** wanghua has joined #openstack-nova | 01:13 | |
| *** zzzeek has joined #openstack-nova | 01:14 | |
| *** thorst1 has joined #openstack-nova | 01:15 | |
| *** thorst1 has quit IRC | 01:23 | |
| openstackgerrit | Li Yuanyuan proposed openstack/nova: Fix exception message mistake https://review.openstack.org/215509 | 01:25 |
| *** artom has quit IRC | 01:28 | |
| *** humble_ has joined #openstack-nova | 01:32 | |
| *** meghal has joined #openstack-nova | 01:33 | |
| *** meghal1 has joined #openstack-nova | 01:33 | |
| openstackgerrit | Li Yuanyuan proposed openstack/nova: Fix exception message mistake https://review.openstack.org/215509 | 01:34 |
| *** Guest62749 has joined #openstack-nova | 01:34 | |
| *** wanghua has quit IRC | 01:35 | |
| *** y_sawai has joined #openstack-nova | 01:35 | |
| *** signed8bit has quit IRC | 01:35 | |
| *** Guest62749 has quit IRC | 01:36 | |
| *** meghal has quit IRC | 01:37 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: DB:support for migration objects multi-statuses https://review.openstack.org/169604 | 01:39 |
| openstackgerrit | Li Yuanyuan proposed openstack/nova: Fix exception message mistake https://review.openstack.org/215509 | 01:42 |
| *** penick has joined #openstack-nova | 01:49 | |
| *** shz has quit IRC | 01:50 | |
| *** liusheng has quit IRC | 01:50 | |
| *** liusheng has joined #openstack-nova | 01:50 | |
| *** hanzhang has joined #openstack-nova | 01:50 | |
| *** lyanchih has joined #openstack-nova | 01:51 | |
| *** tjones has joined #openstack-nova | 01:56 | |
| *** penick has quit IRC | 01:59 | |
| mriedem | mtreinish: isn't it nice to drop a -1 on that many +1s? https://review.openstack.org/#/c/214920/ | 02:04 |
| mriedem | dansmith: ^ | 02:04 |
| *** htruta has quit IRC | 02:05 | |
| *** dims has quit IRC | 02:06 | |
| *** JoshNang has quit IRC | 02:06 | |
| *** jwcroppe has joined #openstack-nova | 02:07 | |
| *** lyanchih has quit IRC | 02:08 | |
| *** htruta has joined #openstack-nova | 02:09 | |
| *** JoshNang has joined #openstack-nova | 02:09 | |
| *** boris-42 has quit IRC | 02:10 | |
| *** jwcroppe has quit IRC | 02:11 | |
| *** lyanchih has joined #openstack-nova | 02:12 | |
| *** unicell1 has quit IRC | 02:13 | |
| *** SourabhP has joined #openstack-nova | 02:13 | |
| *** asettle is now known as asettle-afk | 02:13 | |
| *** asettle-afk has quit IRC | 02:13 | |
| *** SourabhP_ has joined #openstack-nova | 02:16 | |
| *** SourabhP has quit IRC | 02:19 | |
| *** SourabhP_ is now known as SourabhP | 02:19 | |
| *** yamahata has quit IRC | 02:19 | |
| *** Apoorva has quit IRC | 02:19 | |
| *** raginbajin has quit IRC | 02:23 | |
| *** tjones has left #openstack-nova | 02:23 | |
| *** haomaiwang has joined #openstack-nova | 02:23 | |
| *** haomaiwang has quit IRC | 02:24 | |
| gmann | alex_xu: johnthetubaguy : added VIF extension thing on https://etherpad.openstack.org/p/liberty-nova-priorities-tracking | 02:25 |
| alex_xu | gmann: I think we are late for that | 02:25 |
| gmann | mriedem: please check in your free time - https://review.openstack.org/#/c/198944/ | 02:25 |
| *** haomaiwang has joined #openstack-nova | 02:25 | |
| gmann | alex_xu: that is bug fix which spec is already merged | 02:25 |
| alex_xu | gmann: yea, we can try. Actually the most important part(the extension list) fix already merged :) | 02:26 |
| gmann | alex_xu: yea that was imp. me also fine if we have review less bandwidth now :) | 02:27 |
| *** raginbajin has joined #openstack-nova | 02:28 | |
| *** achanda has joined #openstack-nova | 02:28 | |
| alex_xu | gmann: yea, let's try | 02:28 |
| gmann | alex_xu: +1 | 02:28 |
| gmann | alex_xu: regarding 500 -> 400 | 02:28 |
| *** READ10 has joined #openstack-nova | 02:29 | |
| gmann | alex_xu: https://review.openstack.org/#/c/217577/2/doc/source/api_microversion_dev.rst | 02:29 |
| gmann | alex_xu: should that be valid for delete and get method also. please check my comments | 02:29 |
| gmann | alex_xu: or something m missing here :) | 02:29 |
| alex_xu | gmann: emm...good point | 02:29 |
| alex_xu | gmann: but I think we still have other case will return 400 | 02:30 |
| alex_xu | gmann: for example, invalid API Version | 02:30 |
| gmann | alex_xu: personally says i like 500 fix to 400 should not require microversion as mriedem does :) but we need to find some good reason | 02:30 |
| gmann | alex_xu: ohh | 02:30 |
| alex_xu | gmann: yea, agree we need good reason | 02:31 |
| alex_xu | gmann: https://github.com/openstack/nova/blob/master/nova/api/openstack/wsgi.py#L746 | 02:31 |
| alex_xu | gmann: maybe this also for this case https://github.com/openstack/nova/blob/master/nova/api/openstack/wsgi.py#L821 | 02:32 |
| mriedem | alex_xu: if an api doesn't have a rule in the policy.json http://git.openstack.org/cgit/openstack/nova/tree/etc/nova/policy.json#n438 | 02:33 |
| *** dims has joined #openstack-nova | 02:33 | |
| mriedem | does it default to 'default'? | 02:33 |
| mriedem | which is admin_or_owner | 02:33 |
| alex_xu | mriedem: yes, default to 'default' | 02:33 |
| mriedem | ok, i was confused b/c there are a lot of apis that also explicitly map to rule:admin_or_owner | 02:33 |
| gmann | alex_xu: ahh i see. Yea those are for all APi request | 02:34 |
| gmann | alex_xu: Thanks | 02:34 |
| alex_xu | gmann: np | 02:34 |
| alex_xu | mriedem: what is 'b/c'? | 02:34 |
| mriedem | because | 02:34 |
| mriedem | i was looking at this https://review.openstack.org/#/c/193117/ | 02:35 |
| alex_xu | mriedem: ah, I see now, thanks | 02:35 |
| *** mrodden has quit IRC | 02:35 | |
| mriedem | and wanted to make sure that change didn't prevent the admin from doing something on a non-admin's server groups | 02:35 |
| *** ctlaugh has quit IRC | 02:35 | |
| alex_xu | mriedem: I guess won't, but let me double check | 02:36 |
| mriedem | it doesn't look like it will, the project_only flag in the db api model_query is only enforced if context is passed and it's a user context (not admin) | 02:37 |
| mriedem | in which case it filters the results by project_id | 02:37 |
| alex_xu | mriedem: yea, the model_query will ignore project_only flag when admin requested | 02:37 |
| *** mrodden has joined #openstack-nova | 02:38 | |
| *** shuquan has joined #openstack-nova | 02:39 | |
| *** ctlaugh has joined #openstack-nova | 02:39 | |
| *** penick has joined #openstack-nova | 02:40 | |
| *** shyu has joined #openstack-nova | 02:41 | |
| *** summerWendy has joined #openstack-nova | 02:42 | |
| *** dave-mccowan has quit IRC | 02:43 | |
| *** haomaiwang has quit IRC | 02:44 | |
| *** summer has quit IRC | 02:44 | |
| openstackgerrit | Huang Zhiteng proposed openstack/nova: libvirt: Force iSCSI disconnect after timeout https://review.openstack.org/167815 | 02:45 |
| *** agireud has joined #openstack-nova | 02:46 | |
| *** gyee has quit IRC | 02:51 | |
| *** SourabhP has quit IRC | 02:52 | |
| *** hakimo has joined #openstack-nova | 02:52 | |
| *** penick has quit IRC | 02:54 | |
| *** shoutm has quit IRC | 02:54 | |
| *** hakimo_ has quit IRC | 02:54 | |
| *** shuquan has quit IRC | 02:57 | |
| *** haomaiwang has joined #openstack-nova | 02:58 | |
| *** READ10 has quit IRC | 02:59 | |
| *** haomaiwang has quit IRC | 03:01 | |
| *** baoli has joined #openstack-nova | 03:01 | |
| *** haomaiwang has joined #openstack-nova | 03:02 | |
| *** shuquan has joined #openstack-nova | 03:03 | |
| *** raildo-afk has quit IRC | 03:05 | |
| *** mriedem has quit IRC | 03:06 | |
| *** browne has joined #openstack-nova | 03:06 | |
| *** asettle has joined #openstack-nova | 03:07 | |
| *** raildo-afk has joined #openstack-nova | 03:08 | |
| *** lyanchih has quit IRC | 03:08 | |
| *** lyanchih has joined #openstack-nova | 03:11 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add UniqueConstraint for quota usages table https://review.openstack.org/212908 | 03:11 |
| *** nelsnels_ has quit IRC | 03:17 | |
| *** nelsnelson has joined #openstack-nova | 03:18 | |
| *** haomaiwang has quit IRC | 03:18 | |
| *** shuquan has quit IRC | 03:18 | |
| *** dims has quit IRC | 03:23 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova-specs: Add project id and user id information in os-server-group API https://review.openstack.org/209917 | 03:25 |
| *** BadCub has quit IRC | 03:27 | |
| *** swamireddy has quit IRC | 03:27 | |
| *** meghal1 has quit IRC | 03:27 | |
| *** haomaiwang has joined #openstack-nova | 03:27 | |
| *** meghal has joined #openstack-nova | 03:28 | |
| *** comstud has quit IRC | 03:28 | |
| *** jroll has quit IRC | 03:28 | |
| *** Vek has quit IRC | 03:28 | |
| *** mgagne has quit IRC | 03:28 | |
| *** d34dh0r53 has quit IRC | 03:28 | |
| *** lbragstad has quit IRC | 03:28 | |
| *** wendar has quit IRC | 03:28 | |
| *** toan has quit IRC | 03:28 | |
| *** gus has quit IRC | 03:28 | |
| *** Nakato has quit IRC | 03:28 | |
| *** wendar_ has joined #openstack-nova | 03:28 | |
| *** shuquan has joined #openstack-nova | 03:29 | |
| *** reed has quit IRC | 03:29 | |
| *** apmelton has quit IRC | 03:29 | |
| *** swamireddy has joined #openstack-nova | 03:29 | |
| *** salv-orlando has joined #openstack-nova | 03:30 | |
| *** toan has joined #openstack-nova | 03:30 | |
| *** reed has joined #openstack-nova | 03:31 | |
| *** deepthi has joined #openstack-nova | 03:31 | |
| *** BadCub has joined #openstack-nova | 03:31 | |
| *** Nakato has joined #openstack-nova | 03:31 | |
| *** gus has joined #openstack-nova | 03:31 | |
| *** BadCub has quit IRC | 03:31 | |
| *** apmelton has joined #openstack-nova | 03:32 | |
| *** d34dh0r53 has joined #openstack-nova | 03:32 | |
| *** jroll has joined #openstack-nova | 03:32 | |
| *** tojuvone has joined #openstack-nova | 03:32 | |
| *** agireud has quit IRC | 03:33 | |
| *** fawadkhaliq has joined #openstack-nova | 03:34 | |
| *** mgagne has joined #openstack-nova | 03:34 | |
| *** haomaiwang has quit IRC | 03:36 | |
| *** salv-orlando has quit IRC | 03:36 | |
| *** comstud has joined #openstack-nova | 03:37 | |
| *** lbragstad has joined #openstack-nova | 03:37 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Pass block_device_info when delete an encrypted lvm https://review.openstack.org/216123 | 03:37 |
| *** summerWendy has quit IRC | 03:39 | |
| *** yuntongjin1 has quit IRC | 03:39 | |
| *** kun_huang has joined #openstack-nova | 03:40 | |
| kun_huang | guys, how to trigger abandon on the patch with no response to -1? | 03:41 |
| *** achanda has quit IRC | 03:41 | |
| RuiChen | hi mriedem , I will update the patch https://review.openstack.org/#/c/193117/ by follow your comments, thank you. | 03:47 |
| RuiChen | and can you take a look this patch again https://review.openstack.org/#/c/211083/ , 1:M mapping still exist in Ironic driver. | 03:49 |
| *** baoli has quit IRC | 03:57 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add checks for volume status when rebuilding https://review.openstack.org/217567 | 03:58 |
| *** jwcroppe has joined #openstack-nova | 03:59 | |
| *** zzzeek has quit IRC | 03:59 | |
| *** unicell has joined #openstack-nova | 04:04 | |
| *** liusheng has quit IRC | 04:05 | |
| *** hemna has joined #openstack-nova | 04:05 | |
| *** armax has quit IRC | 04:09 | |
| *** rajesht has joined #openstack-nova | 04:09 | |
| *** links has joined #openstack-nova | 04:09 | |
| *** tpeoples has joined #openstack-nova | 04:09 | |
| *** rajesht has quit IRC | 04:10 | |
| *** tjones has joined #openstack-nova | 04:10 | |
| *** rajesht has joined #openstack-nova | 04:10 | |
| *** shuquan has quit IRC | 04:13 | |
| *** armax has joined #openstack-nova | 04:16 | |
| *** tjones has left #openstack-nova | 04:26 | |
| *** haomaiwang has joined #openstack-nova | 04:26 | |
| *** vilobhmm1 has joined #openstack-nova | 04:27 | |
| *** frickler has quit IRC | 04:30 | |
| *** BadCub has joined #openstack-nova | 04:32 | |
| *** lyanchih has quit IRC | 04:33 | |
| *** haomaiwang has quit IRC | 04:50 | |
| *** yuntongjin has joined #openstack-nova | 04:52 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add checks for volume status when rebuilding https://review.openstack.org/217567 | 04:53 |
| *** oomichi has quit IRC | 04:53 | |
| *** meghal has quit IRC | 04:55 | |
| *** deepakcs has joined #openstack-nova | 04:56 | |
| deepakcs | Hello cores, can someone provide a +A to https://review.openstack.org/#/c/212518/ (It already has +2 from danpb) | 04:56 |
| *** hparekh2 has joined #openstack-nova | 05:05 | |
| *** lyanchih has joined #openstack-nova | 05:06 | |
| *** hparekh has quit IRC | 05:06 | |
| *** meghal has joined #openstack-nova | 05:07 | |
| *** hparekh has joined #openstack-nova | 05:09 | |
| *** hparekh2 has quit IRC | 05:10 | |
| *** ijw has quit IRC | 05:12 | |
| *** ijw has joined #openstack-nova | 05:12 | |
| *** ijw has quit IRC | 05:14 | |
| *** ijw has joined #openstack-nova | 05:14 | |
| *** hemna has quit IRC | 05:15 | |
| *** ildikov has joined #openstack-nova | 05:19 | |
| *** shuquan has joined #openstack-nova | 05:23 | |
| vishwanathj | hi, I am looking to download a cirros image that is cloud-init enabled, would appreciate if you folks can share a download link, I would like to use this image to test out the user_data property in heat | 05:24 |
| *** summer has joined #openstack-nova | 05:24 | |
| *** nelsnelson has quit IRC | 05:25 | |
| *** Vek has joined #openstack-nova | 05:28 | |
| *** shuquan has quit IRC | 05:30 | |
| *** salv-orlando has joined #openstack-nova | 05:31 | |
| *** haomaiwang has joined #openstack-nova | 05:33 | |
| *** agireud has joined #openstack-nova | 05:33 | |
| *** haomaiwang has quit IRC | 05:35 | |
| *** salv-orlando has quit IRC | 05:36 | |
| openstackgerrit | javeme proposed openstack/nova: VMware: Update to return the correct ESX iqn https://review.openstack.org/162700 | 05:36 |
| openstackgerrit | lyanchih proposed openstack/nova: Remove fixed test in tempest-dsvm-cells-rc https://review.openstack.org/205512 | 05:37 |
| openstackgerrit | lyanchih proposed openstack/nova: Creating instance fail when inject ssh key in cells mode https://review.openstack.org/197514 | 05:37 |
| *** jwcroppe has quit IRC | 05:41 | |
| *** meghal has quit IRC | 05:41 | |
| *** jwcroppe has joined #openstack-nova | 05:42 | |
| *** meghal has joined #openstack-nova | 05:42 | |
| *** skylerberg has quit IRC | 05:47 | |
| *** agireud has quit IRC | 05:47 | |
| *** jwcroppe has quit IRC | 05:48 | |
| *** yamahata has joined #openstack-nova | 05:52 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Add warning log when deprecated v2 and v3 code get used https://review.openstack.org/218103 | 05:53 |
| alex_xu | johnthetubaguy: sdague ^ | 05:55 |
| *** armax has quit IRC | 05:56 | |
| *** shyu has quit IRC | 05:56 | |
| openstackgerrit | Feodor Tersin proposed openstack/nova: DNM Adding logging around bdm actions for cells https://review.openstack.org/217829 | 06:00 |
| *** Marga_ has joined #openstack-nova | 06:00 | |
| *** Marga_ has quit IRC | 06:00 | |
| *** Marga_ has joined #openstack-nova | 06:01 | |
| *** haomaiwang has joined #openstack-nova | 06:02 | |
| *** yuntongjin has quit IRC | 06:05 | |
| *** yuntongjin has joined #openstack-nova | 06:06 | |
| *** asettle has quit IRC | 06:07 | |
| *** shuquan has joined #openstack-nova | 06:14 | |
| *** yuntongjin has quit IRC | 06:16 | |
| *** salv-orlando has joined #openstack-nova | 06:18 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add UniqueConstraint for quota usages table https://review.openstack.org/212908 | 06:18 |
| *** hparekh has quit IRC | 06:20 | |
| *** hparekh has joined #openstack-nova | 06:20 | |
| *** tpeoples has quit IRC | 06:20 | |
| *** Sree has joined #openstack-nova | 06:21 | |
| *** haomaiwang has quit IRC | 06:23 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Imported Translations from Transifex https://review.openstack.org/212367 | 06:23 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add checks for volume status when rebuilding https://review.openstack.org/217567 | 06:23 |
| *** shuquan has quit IRC | 06:25 | |
| rajesht | alaski, dansmith: you around ? | 06:25 |
| *** shuquan has joined #openstack-nova | 06:26 | |
| *** shuquan has quit IRC | 06:26 | |
| *** shuquan has joined #openstack-nova | 06:27 | |
| *** fawadkhaliq has quit IRC | 06:29 | |
| *** andrearosa has quit IRC | 06:31 | |
| *** andrearosa has joined #openstack-nova | 06:31 | |
| *** shuquan has quit IRC | 06:32 | |
| *** agireud has joined #openstack-nova | 06:33 | |
| *** agireud has quit IRC | 06:38 | |
| openstackgerrit | huan proposed openstack/nova: Do not rely on XenAPI to guess device names https://review.openstack.org/213112 | 06:39 |
| *** jwcroppe has joined #openstack-nova | 06:44 | |
| *** yuntongjin has joined #openstack-nova | 06:45 | |
| *** nkrinner has joined #openstack-nova | 06:48 | |
| *** vivekd has joined #openstack-nova | 06:48 | |
| *** shuquan has joined #openstack-nova | 06:54 | |
| openstackgerrit | Rui Chen proposed openstack/nova: Fix permission issue of server group API https://review.openstack.org/193117 | 06:55 |
| *** liusheng has joined #openstack-nova | 07:02 | |
| *** vilobhmm1 has quit IRC | 07:02 | |
| *** yuntongjin1 has joined #openstack-nova | 07:02 | |
| *** yuntongjin has quit IRC | 07:04 | |
| *** achanda has joined #openstack-nova | 07:05 | |
| *** Nikolay_St has joined #openstack-nova | 07:05 | |
| *** Kevin_Zheng has quit IRC | 07:05 | |
| *** meghal has quit IRC | 07:09 | |
| *** yuntongjin has joined #openstack-nova | 07:12 | |
| *** stefan_amann has joined #openstack-nova | 07:12 | |
| *** EinstCrazy has quit IRC | 07:13 | |
| *** yuntongjin1 has quit IRC | 07:14 | |
| *** zhangjn has quit IRC | 07:15 | |
| *** browne has quit IRC | 07:17 | |
| *** jwcroppe has quit IRC | 07:18 | |
| *** fawadkhaliq has joined #openstack-nova | 07:18 | |
| *** mpaolino has joined #openstack-nova | 07:19 | |
| *** yuntongjin1 has joined #openstack-nova | 07:21 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add checks for volume status when rebuilding https://review.openstack.org/217567 | 07:21 |
| openstackgerrit | Alexis Lee proposed openstack/nova: Wait for device to be mapped https://review.openstack.org/215613 | 07:22 |
| *** yuntongjin has quit IRC | 07:23 | |
| *** summer has quit IRC | 07:30 | |
| *** frickler has joined #openstack-nova | 07:32 | |
| *** Marga_ has quit IRC | 07:34 | |
| *** Marga_ has joined #openstack-nova | 07:35 | |
| *** haomaiwang has joined #openstack-nova | 07:38 | |
| *** jichen has joined #openstack-nova | 07:43 | |
| *** haomaiwang has quit IRC | 07:44 | |
| *** haomaiwa_ has joined #openstack-nova | 07:45 | |
| *** mpaolino has quit IRC | 07:48 | |
| *** shuquan_ has joined #openstack-nova | 07:49 | |
| *** haomaiw__ has joined #openstack-nova | 07:49 | |
| *** shuquan has quit IRC | 07:50 | |
| *** shuquan_ is now known as shuquan | 07:50 | |
| *** haomaiwa_ has quit IRC | 07:50 | |
| *** ndipanov has joined #openstack-nova | 07:51 | |
| *** ndipanov has quit IRC | 07:52 | |
| *** doude has joined #openstack-nova | 07:52 | |
| *** ndipanov has joined #openstack-nova | 07:52 | |
| *** fawadk has joined #openstack-nova | 07:53 | |
| *** liusheng has quit IRC | 07:53 | |
| *** fawadkhaliq has quit IRC | 07:53 | |
| *** liusheng has joined #openstack-nova | 07:54 | |
| *** matrohon has joined #openstack-nova | 07:55 | |
| *** alex_klimov has joined #openstack-nova | 07:56 | |
| *** achanda has quit IRC | 07:56 | |
| *** haomaiw__ has quit IRC | 07:58 | |
| *** markus_z has joined #openstack-nova | 07:58 | |
| *** shuquan_ has joined #openstack-nova | 08:00 | |
| *** shuquan has quit IRC | 08:00 | |
| *** shuquan_ is now known as shuquan | 08:00 | |
| *** alex_klimov has quit IRC | 08:01 | |
| *** alex_klimov has joined #openstack-nova | 08:02 | |
| bauzas | BobBall-AWOL: heya, looking at http://zuul.openstack.xenproject.org/scoreboard/?project=openstack/nova&user=jenkins,citrix_xenserver_ci,XenProject-CI&timeframe=24 it seems that XenServer CI is disabled, right? | 08:04 |
| *** leopoldj has joined #openstack-nova | 08:04 | |
| bauzas | BobBall-AWOL: since I have a couple of negative votes, I can't then recheck, right? | 08:05 |
| *** markus_z has quit IRC | 08:06 | |
| *** yassine__ has joined #openstack-nova | 08:08 | |
| *** danpb has joined #openstack-nova | 08:10 | |
| *** tojuvone has quit IRC | 08:10 | |
| *** derekh has joined #openstack-nova | 08:10 | |
| *** jistr has joined #openstack-nova | 08:12 | |
| *** Sree has quit IRC | 08:15 | |
| *** yamahata has quit IRC | 08:17 | |
| *** bfic has joined #openstack-nova | 08:18 | |
| johnthetubaguy | bauzas: I think Bob is on holiday now | 08:23 |
| bauzas | johnthetubaguy: okay, I just wonder how I can clean that up | 08:24 |
| johnthetubaguy | bauzas: rebase ;) | 08:24 |
| bauzas | johnthetubaguy: because Gerrit shows -1 for my changes | 08:24 |
| bauzas | johnthetubaguy: urgh | 08:24 |
| johnthetubaguy | bauzas: did recheck not work? | 08:24 |
| bauzas | johnthetubaguy: for xenserver, nope because it seems it's disabled as per http://zuul.openstack.xenproject.org/scoreboard/?project=openstack/nova&user=jenkins,citrix_xenserver_ci,XenProject-CI&timeframe=24 | 08:24 |
| *** lucasagomes has joined #openstack-nova | 08:25 | |
| johnthetubaguy | bauzas: I would try a recheck, its possible the tooling is just bad | 08:25 |
| *** IanGovett has joined #openstack-nova | 08:25 | |
| bauzas | johnthetubaguy: I tried a couple of times last night :/ | 08:25 |
| johnthetubaguy | oh, OK, I guess its totally broken then | 08:25 |
| *** rdopiera has joined #openstack-nova | 08:25 | |
| bauzas | johnthetubaguy: but I can spin a rebase for the allocation-ratio series | 08:25 |
| johnthetubaguy | I know he tried to disable a test, then restart it, seems like it didn't come back up | 08:25 |
| *** rdopiera has quit IRC | 08:25 | |
| *** rdopiera has joined #openstack-nova | 08:26 | |
| BobBall-AWOL | sorry bauzas | 08:26 |
| BobBall-AWOL | I'm restarting now | 08:26 |
| johnthetubaguy | bauzas: yeah, I wouldn't worry about the -1 really | 08:26 |
| bauzas | johnthetubaguy: I'm tho worried by the ReqSpec series | 08:26 |
| johnthetubaguy | BobBall-AWOL: cool thanks | 08:26 |
| BobBall-AWOL | zuul took forever to merge the change to xenapi-os-testing | 08:26 |
| BobBall-AWOL | (which disabled the test) | 08:26 |
| bauzas | BobBall-AWOL: oh coolness, beer_count['BobBall']++ | 08:26 |
| johnthetubaguy | ah, bummer | 08:26 |
| johnthetubaguy | yeah, gate is kinda busy | 08:26 |
| BobBall-AWOL | We have a backlog of 80 tests now | 08:27 |
| BobBall-AWOL | but it will work through them... | 08:27 |
| BobBall-AWOL | :) | 08:27 |
| bauzas | johnthetubaguy: the ReqSpec series is kinda hard to maintain because of the number of changes vs. the merge status tbh | 08:27 |
| johnthetubaguy | bauzas: yeah, thats cool | 08:27 |
| *** Sree has joined #openstack-nova | 08:27 | |
| bauzas | BobBall-AWOL: cool, no need to recheck then ? | 08:27 |
| johnthetubaguy | I need to go jump on those | 08:27 |
| BobBall-AWOL | patchset number? | 08:28 |
| bauzas | BobBall-AWOL: well, a couple of them :) | 08:28 |
| *** rdopiera has quit IRC | 08:28 | |
| johnthetubaguy | (most of the 80, giggles) | 08:28 |
| bauzas | BobBall-AWOL: lemme rather do a Gerrit query :) | 08:28 |
| *** rdopiera has joined #openstack-nova | 08:28 | |
| BobBall-AWOL | Just one patchset would be useful | 08:28 |
| BobBall-AWOL | Will check if it's queued | 08:29 |
| *** lyanchih has quit IRC | 08:29 | |
| bauzas | BobBall-AWOL: ack | 08:29 |
| bauzas | BobBall-AWOL: https://review.openstack.org/#/c/188486/30 for example | 08:29 |
| bauzas | BobBall-AWOL: xenserver:recheck done at 10:40UTC | 08:30 |
| BobBall-AWOL | root@dev-ci-controller:~# osci-view list | grep 188486 | 08:30 |
| BobBall-AWOL | | 10111 | openstack/nova | refs/changes/86/188486/30 | Queued | (None) | None | 9.83 | - | | 08:30 |
| *** fawadk has quit IRC | 08:30 | |
| BobBall-AWOL | root@dev-ci-controller:~# osci-view list | grep 188486 | 08:30 |
| BobBall-AWOL | | 10111 | openstack/nova | refs/changes/86/188486/30 | Queued | (None) | None | 9.83 | - | | 08:30 |
| BobBall-AWOL | You're in luck - it's queued! | 08:30 |
| BobBall-AWOL | yay. Give it time. | 08:30 |
| bauzas | alleluiah | 08:30 |
| bauzas | BobBall-AWOL: coolness | 08:30 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add project-id when list server-groups https://review.openstack.org/209018 | 08:30 |
| bauzas | johnthetubaguy: honestly, I think I need alaski's feedback for the Spec object | 08:31 |
| bauzas | johnthetubaguy: if you volunteer, I'd rather ask to to peek at https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/allocation-ratio-to-resource-tracker,n,z | 08:31 |
| johnthetubaguy | bauzas: yep, totally needs his vote on there | 08:31 |
| bauzas | we're still on time, I'm not that worried | 08:32 |
| bauzas | johnthetubaguy: but the RPC change is unfortunately way far in the ReqSpec series | 08:33 |
| *** kun_huang has left #openstack-nova | 08:33 | |
| bauzas | johnthetubaguy: which means that the whole series could be hard for getting a FFE | 08:33 |
| bauzas | whatever the exception process is | 08:33 |
| johnthetubaguy | bauzas: lets not try worry about that right now, lets try get it merged | 08:33 |
| bauzas | agreed | 08:33 |
| *** lpetrut has joined #openstack-nova | 08:34 | |
| bauzas | johnthetubaguy: on other sched bits, some stuff is being done on tests refactoring by jay and some changes are there for the resource-objects BP (the one for objectifying what we send to ComputeNodes) | 08:35 |
| bauzas | johnthetubaguy: FYI | 08:35 |
| bauzas | johnthetubaguy: I'm currently reviewing jay's patches | 08:36 |
| johnthetubaguy | OK, its getting crazy close to the wire, like thing that are not in review now, are really unlikely to merge | 08:36 |
| johnthetubaguy | but you know all that | 08:36 |
| bauzas | yup | 08:37 |
| openstackgerrit | Yixing proposed openstack/nova: Replace except Exception with specific exception https://review.openstack.org/217067 | 08:37 |
| *** lpetrut has left #openstack-nova | 08:39 | |
| bauzas | johnthetubaguy: all of the stuff is tracked in the etherpad of doom | 08:39 |
| *** bfic has quit IRC | 08:40 | |
| *** dbite has quit IRC | 08:41 | |
| *** bfic has joined #openstack-nova | 08:41 | |
| *** dguitarbite has joined #openstack-nova | 08:41 | |
| *** Savemech has quit IRC | 08:42 | |
| *** markus_z has joined #openstack-nova | 08:42 | |
| *** bfic1 has joined #openstack-nova | 08:42 | |
| *** Savemech has joined #openstack-nova | 08:43 | |
| *** bfic has quit IRC | 08:45 | |
| *** lyanchih has joined #openstack-nova | 08:47 | |
| *** shuquan has quit IRC | 08:51 | |
| *** atuvenie has joined #openstack-nova | 08:52 | |
| johnthetubaguy | bauzas: nod (and smile) | 08:54 |
| *** obondarev has quit IRC | 08:57 | |
| *** shuquan has joined #openstack-nova | 08:58 | |
| *** fawadkhaliq has joined #openstack-nova | 08:58 | |
| *** marzif has joined #openstack-nova | 08:59 | |
| *** pkholkin has quit IRC | 08:59 | |
| *** tdurakov has quit IRC | 08:59 | |
| *** fawadk has joined #openstack-nova | 09:00 | |
| *** VikasC has joined #openstack-nova | 09:01 | |
| *** fawadkhaliq has quit IRC | 09:02 | |
| *** exploreshaifali has joined #openstack-nova | 09:04 | |
| bauzas | ndipanov: I'm about to rebase https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:pci-cleanup,n,z , it would be nice if you could then look at my 3-way merge I'm doing in the next iteration of https://review.openstack.org/#/c/213592/6 | 09:04 |
| bauzas | ndipanov: because you're also changing the RT unittests and I would love to see some sync between you and jay here | 09:04 |
| ndipanov | bauzas, not sure what you mean | 09:05 |
| bauzas | ndipanov: np, just saying that since https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/migration-fix-resource-tracking,n,z is merged for a couple of changes, it impacts https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:pci-cleanup,n,z | 09:06 |
| *** snikitin has quit IRC | 09:06 | |
| bauzas | ndipanov: where the latter aims to remove all of the silly test_resource_tracker module | 09:06 |
| *** shuquan_ has joined #openstack-nova | 09:07 | |
| bauzas | ndipanov: so since you're quite changing the RT tests because you're modifying the Claims unittests, I just want to make sure that jay's changes are reflecting the changes you made but within test_tracker | 09:07 |
| *** eglynn has joined #openstack-nova | 09:07 | |
| bauzas | ndipanov: I was not sure you were aware of that series, hence my ping | 09:08 |
| *** shuquan has quit IRC | 09:08 | |
| *** shuquan_ is now known as shuquan | 09:08 | |
| ndipanov | I was not | 09:08 |
| ndipanov | but I've kind of sort of given up hope on that merging in L | 09:08 |
| ndipanov | my stuff I mean | 09:08 |
| ndipanov | I'm too old for that shit | 09:08 |
| ndipanov | I'll just post it again in M | 09:08 |
| ndipanov | so whatever | 09:09 |
| *** y_sawai has quit IRC | 09:09 | |
| ndipanov | johnthetubaguy, I'd like to have the DB changes merged in L if possible for my fix migration RTracking | 09:10 |
| ndipanov | if I pull them to the bottom of the set | 09:10 |
| ndipanov | can we consider them as a partial exception or something? | 09:10 |
| bauzas | ndipanov: well, let's see what we can do | 09:11 |
| bauzas | ndipanov: I just want to make sure that we don't regress because we're merging 2 things at the same time | 09:11 |
| *** lyanchih has quit IRC | 09:11 | |
| ndipanov | bauzas, nah I'm not doing that crunch crap again - done it 2 times and started literally hating everything about nova for months after that | 09:11 |
| *** marzif has quit IRC | 09:12 | |
| *** shuquan has quit IRC | 09:12 | |
| *** marzif has joined #openstack-nova | 09:12 | |
| johnthetubaguy | ndipanov: its an important bug fix, so I can see an argument for it | 09:13 |
| ndipanov | johnthetubaguy, ok I'll pull the changes now and post them so they can get reviews | 09:13 |
| ndipanov | and then let's see come freeze time | 09:13 |
| *** lyanchih has joined #openstack-nova | 09:14 | |
| johnthetubaguy | bauzas: what do you see as the conflict with the migration stuff? | 09:14 |
| ndipanov | johnthetubaguy, tests | 09:14 |
| ndipanov | with jay's set | 09:14 |
| bauzas | johnthetubaguy: https://review.openstack.org/#/c/213589/5 | 09:14 |
| *** Sree has quit IRC | 09:14 | |
| ndipanov | as he is moving all the tests from underneath my set | 09:14 |
| bauzas | johnthetubaguy: I'm not worried by the merge conflict | 09:14 |
| bauzas | johnthetubaguy: I'm more worried by loosing some coverage there | 09:15 |
| bauzas | if we don't verify | 09:15 |
| tobasco | i noticed my compute nodes does not cache images in /var/lib/nova/images, i have a very minimal config; do i have to enable it in config somewhere? | 09:15 |
| bauzas | that's not super difficult, just a comment | 09:15 |
| *** summer has joined #openstack-nova | 09:16 | |
| johnthetubaguy | ndipanov: oh right, that figures I guess | 09:16 |
| ndipanov | fwiw - I see no reason why Jay's set needs to be impacted by the FF | 09:17 |
| ndipanov | it has no BP related anyway | 09:17 |
| ndipanov | and is just test refactoring | 09:17 |
| johnthetubaguy | ndipanov: yeah, test refactoring can be post FF | 09:17 |
| ndipanov | I'll comment | 09:17 |
| johnthetubaguy | it distracts from getting bug fixes in, which sucks, but it might make sense to hold off on that for a few days | 09:17 |
| *** gibi has joined #openstack-nova | 09:18 | |
| johnthetubaguy | yeah, lets try catch jaypipes when he is online later | 09:18 |
| johnthetubaguy | ndipanov: I would rather have that test refactor after FF, than not merge your migration fixes | 09:18 |
| *** shuquan has joined #openstack-nova | 09:18 | |
| johnthetubaguy | (although I thought there was some stuff on top of that we want, but lets ignore that for now) | 09:19 |
| bauzas | johnthetubaguy: agreed with you, but actually even ndipanov's changes are bugfixes IMHO | 09:19 |
| ndipanov | well | 09:19 |
| bauzas | johnthetubaguy: at least the first few of them | 09:19 |
| ndipanov | if we manage to get the DB stuff in | 09:19 |
| johnthetubaguy | bauzas: yeah, I am totally leaning that way | 09:19 |
| ndipanov | then yes the rest could be seen as bugfixing | 09:20 |
| bauzas | ndipanov: well, the DB stuff is a feature, I agree | 09:20 |
| johnthetubaguy | ndipanov: ah, now I see your thinking, that sounds like a good plan | 09:20 |
| bauzas | ndipanov: but the other changes sound very bugfixing IMHO | 09:20 |
| ndipanov | yes agreed | 09:20 |
| johnthetubaguy | ndipanov: we might want to attach bugs to the later patches, just to keep the release manager less twitchy about them (and just hold back on the log strings too), but lets deal with that later | 09:21 |
| ndipanov | yes there is a bug that I will start attaching | 09:22 |
| ndipanov | https://bugs.launchpad.net/nova/+bug/1417667 | 09:22 |
| openstack | Launchpad bug 1417667 in OpenStack Compute (nova) "migration/evacuation/rebuild/resize of instance with dedicated cpus needs to recalculate cpus on destination" [Medium,In progress] - Assigned to Bart Wensley (bartwensley) | 09:22 |
| *** bauzas is now known as bauwser | 09:23 | |
| johnthetubaguy | ndipanov: perfect, that will stop triggering alarm bells for the release team folks | 09:23 |
| *** takedakn has joined #openstack-nova | 09:23 | |
| ndipanov | cool - let me get the DB stuff sorted | 09:23 |
| johnthetubaguy | its purely a case of tracking doc and string translation impact really, as I understand it | 09:23 |
| johnthetubaguy | ndipanov: sounds good | 09:24 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add checks for volume status when rebuilding https://review.openstack.org/217567 | 09:24 |
| deepakcs | Hello cores, can someone provide a +A to https://review.openstack.org/#/c/212518/ (It already has +2 from danpb) | 09:26 |
| deepakcs | ndipanov: johnthetubaguy ^^ | 09:26 |
| ndipanov | ah yeah I saw that one... | 09:26 |
| ndipanov | I really need to sit down and look at that one as it messes with a tricky bit of code | 09:27 |
| ndipanov | has it been tested deepakcs ? | 09:27 |
| deepakcs | ndipanov: yes and it has a testcase too | 09:27 |
| deepakcs | ndipanov: no, the change is just 2 lines, 'guess ur confusing it with somethign else ? | 09:27 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: rt: Rewrite abort and update_usage tests https://review.openstack.org/213590 | 09:28 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Move test of claim context manager to test_tracker https://review.openstack.org/213591 | 09:28 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Cleanup RT _instance_in_resize_state() https://review.openstack.org/213589 | 09:28 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Move resize claim abort test to test_tracker https://review.openstack.org/213592 | 09:28 |
| *** claudiub has joined #openstack-nova | 09:31 | |
| *** takedakn has quit IRC | 09:33 | |
| *** haomaiwang has joined #openstack-nova | 09:34 | |
| *** tdurakov_ has joined #openstack-nova | 09:34 | |
| *** lyanchih has quit IRC | 09:38 | |
| *** tdurakov_ has quit IRC | 09:39 | |
| *** pkholkin has joined #openstack-nova | 09:40 | |
| *** marzif has quit IRC | 09:41 | |
| *** salv-orlando has quit IRC | 09:42 | |
| *** lyanchih has joined #openstack-nova | 09:43 | |
| *** gszasz has joined #openstack-nova | 09:43 | |
| bauwser | ndipanov: could you please remove your -1 on https://review.openstack.org/#/c/216362/5 since I provided what we discussed in a later iteration, btw ? | 09:44 |
| ndipanov | bauwser, looking | 09:45 |
| ndipanov | bauwser, but pleases respond then | 09:45 |
| ndipanov | with details for people who come back to read the discussion | 09:45 |
| bauwser | ndipanov: ack | 09:45 |
| *** snikitin has joined #openstack-nova | 09:47 | |
| *** tdurakov has joined #openstack-nova | 09:48 | |
| *** yuntongjin1 has quit IRC | 09:48 | |
| *** alex_klimov has quit IRC | 09:48 | |
| bauwser | ndipanov: done | 09:49 |
| *** sudipto has joined #openstack-nova | 09:50 | |
| *** summer has quit IRC | 09:51 | |
| sudipto | johnthetubaguy, I was working on a refactor of the glancewrapper in nova to not use the V1 APIs by default. The code is not yet up for review...do you think it's something that should now go for Mitaka? | 09:52 |
| johnthetubaguy | sudipto: there are several others working on that, as I understand it | 09:52 |
| johnthetubaguy | sudipto: the blueprint and spec we had to describe the agreed way forward has been deferred to mitaka | 09:53 |
| *** rushiagr_away is now known as rushiagr | 09:53 | |
| johnthetubaguy | sudipto: we really need focus on bug fixes and reviewing bug fixes at this point plese | 09:53 |
| sudipto | johnthetubaguy, ah ok, I wasn't aware of that. Just had a discussion with Jay before thinking of working on it. | 09:53 |
| sudipto | johnthetubaguy, yeah sure...i didn't start on it - just wanted to double check. | 09:53 |
| johnthetubaguy | sudipto: there have been quite a few competing efforts at this point, I would reach out to nikhil_k about the details of where things are at right now | 09:53 |
| sudipto | johnthetubaguy, alrite sure... | 09:54 |
| johnthetubaguy | sudipto: this was the spec that needs reworking for mitaka, that might be a good place to start: https://review.openstack.org/#/c/194945/ | 09:55 |
| *** alex_klimov has joined #openstack-nova | 09:56 | |
| johnthetubaguy | sudipto: having said all that, I would like to see you an jay help steer that effort in a good direction | 09:56 |
| johnthetubaguy | sudipto: but to answer your original question, totally something for mitaka at this point | 09:56 |
| sudipto | johnthetubaguy, sure...I will have another chat with Jay...but Mitaka it is. The branch will be forked next month i believe? | 09:58 |
| *** bfic1 has quit IRC | 09:58 | |
| *** dave-mccowan has joined #openstack-nova | 09:58 | |
| johnthetubaguy | sudipto: as soon as RC1 gets tagged, yes | 09:58 |
| sudipto | johnthetubaguy, alrite! Thanks for the info! :) | 09:58 |
| johnthetubaguy | sudipto: non-priority feature freeze was 5 weeks ago | 09:59 |
| sudipto | johnthetubaguy, yeah, I know my patches were abandoned for that one :D | 09:59 |
| *** kiran-r has joined #openstack-nova | 09:59 | |
| * sudipto is the non-priority feature guy lol | 09:59 | |
| johnthetubaguy | yeah, we had about 20 or so blueprints hit by that | 10:00 |
| johnthetubaguy | well, more like 30 | 10:00 |
| sudipto | johnthetubaguy, https://review.openstack.org/#/c/213317/ this one - would you recommend a merge? | 10:00 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add checks for volume status when rebuilding https://review.openstack.org/217567 | 10:00 |
| *** haomaiwang has quit IRC | 10:01 | |
| johnthetubaguy | sudipto: bug fixes are good, but I am focusing on priority features that need to merge by tuesday | 10:01 |
| sudipto | johnthetubaguy, alrite sure...i will wait on it. | 10:01 |
| *** bfic has joined #openstack-nova | 10:01 | |
| *** haomaiwa_ has joined #openstack-nova | 10:02 | |
| openstackgerrit | garyk proposed openstack/nova: VMware: create method for getting datacenter from datastore https://review.openstack.org/205819 | 10:03 |
| openstackgerrit | garyk proposed openstack/nova: VMware: prevent volume deletion if volume in instance folder https://review.openstack.org/193103 | 10:04 |
| ftersin | alaski: hi, i've put some info into comments on https://review.openstack.org/#/c/217829/ | 10:06 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Make query to quota usage table order preserved https://review.openstack.org/213584 | 10:07 |
| *** bfic has quit IRC | 10:11 | |
| openstackgerrit | John Garbutt proposed openstack/nova: api: use v2.1 only in api-paste.ini https://review.openstack.org/214085 | 10:14 |
| *** shuquan has quit IRC | 10:19 | |
| *** ildikov has quit IRC | 10:21 | |
| *** jwcroppe has joined #openstack-nova | 10:22 | |
| *** lpetrut1 has joined #openstack-nova | 10:25 | |
| openstackgerrit | John Garbutt proposed openstack/nova: api: deprecate the api v2 extension configuration https://review.openstack.org/214601 | 10:25 |
| *** karimb has joined #openstack-nova | 10:27 | |
| *** haomaiwa_ has quit IRC | 10:27 | |
| *** dguitarbite has quit IRC | 10:27 | |
| *** mgagne has quit IRC | 10:27 | |
| *** apmelton has quit IRC | 10:27 | |
| *** toan has quit IRC | 10:27 | |
| *** swamireddy has quit IRC | 10:27 | |
| *** wendar_ has quit IRC | 10:27 | |
| *** hakimo has quit IRC | 10:27 | |
| *** hanzhang has quit IRC | 10:27 | |
| *** adam_g has quit IRC | 10:27 | |
| *** aix has quit IRC | 10:27 | |
| *** Kennan has quit IRC | 10:27 | |
| *** jgriffith has quit IRC | 10:27 | |
| *** ccarmack has quit IRC | 10:27 | |
| *** jamielennox has quit IRC | 10:27 | |
| *** phschwartz has quit IRC | 10:27 | |
| *** vishwanathj has quit IRC | 10:27 | |
| *** xgsa has quit IRC | 10:27 | |
| *** stefanb has quit IRC | 10:27 | |
| *** sabari has quit IRC | 10:27 | |
| *** sayali has quit IRC | 10:27 | |
| *** flaper87 has quit IRC | 10:27 | |
| *** nikhil_k has quit IRC | 10:27 | |
| *** jogo has quit IRC | 10:27 | |
| *** paboldin has quit IRC | 10:27 | |
| *** lxsli has quit IRC | 10:27 | |
| *** tcammann has quit IRC | 10:27 | |
| *** gilliard has quit IRC | 10:27 | |
| *** tlbr has quit IRC | 10:27 | |
| *** Marga_ has quit IRC | 10:27 | |
| *** Vek has quit IRC | 10:27 | |
| *** BadCub has quit IRC | 10:27 | |
| *** links has quit IRC | 10:27 | |
| *** comstud has quit IRC | 10:27 | |
| *** lbragstad has quit IRC | 10:27 | |
| *** jroll has quit IRC | 10:27 | |
| *** d34dh0r53 has quit IRC | 10:27 | |
| *** gus has quit IRC | 10:27 | |
| *** Nakato has quit IRC | 10:27 | |
| *** markvoelker has quit IRC | 10:27 | |
| *** davideagnello has quit IRC | 10:27 | |
| *** neelabh_ has quit IRC | 10:27 | |
| *** jwang has quit IRC | 10:27 | |
| *** s1rp_ has quit IRC | 10:27 | |
| *** ttx has quit IRC | 10:27 | |
| *** mrsoul has quit IRC | 10:27 | |
| *** tteggel_ has quit IRC | 10:27 | |
| *** leseb has quit IRC | 10:27 | |
| *** _gryf has quit IRC | 10:27 | |
| *** adreznec has quit IRC | 10:27 | |
| *** DinaBelova has quit IRC | 10:27 | |
| *** raorn has quit IRC | 10:27 | |
| *** krotscheck has quit IRC | 10:27 | |
| *** anteaya has quit IRC | 10:27 | |
| *** lpetrut1 is now known as lpetrut | 10:27 | |
| *** Marga_ has joined #openstack-nova | 10:29 | |
| *** Vek has joined #openstack-nova | 10:29 | |
| *** BadCub has joined #openstack-nova | 10:29 | |
| *** links has joined #openstack-nova | 10:29 | |
| *** lbragstad has joined #openstack-nova | 10:29 | |
| *** comstud has joined #openstack-nova | 10:29 | |
| *** jroll has joined #openstack-nova | 10:29 | |
| *** d34dh0r53 has joined #openstack-nova | 10:29 | |
| *** gus has joined #openstack-nova | 10:29 | |
| *** Nakato has joined #openstack-nova | 10:29 | |
| *** markvoelker has joined #openstack-nova | 10:29 | |
| *** davideagnello has joined #openstack-nova | 10:29 | |
| *** neelabh_ has joined #openstack-nova | 10:29 | |
| *** jwang has joined #openstack-nova | 10:29 | |
| *** s1rp_ has joined #openstack-nova | 10:29 | |
| *** ttx has joined #openstack-nova | 10:29 | |
| *** mrsoul has joined #openstack-nova | 10:29 | |
| *** tteggel_ has joined #openstack-nova | 10:29 | |
| *** leseb has joined #openstack-nova | 10:29 | |
| *** _gryf has joined #openstack-nova | 10:29 | |
| *** adreznec has joined #openstack-nova | 10:29 | |
| *** DinaBelova has joined #openstack-nova | 10:29 | |
| *** raorn has joined #openstack-nova | 10:29 | |
| *** krotscheck has joined #openstack-nova | 10:29 | |
| *** anteaya has joined #openstack-nova | 10:29 | |
| *** lpetrut is now known as 18VAAKDGR | 10:29 | |
| *** haomaiwa_ has joined #openstack-nova | 10:29 | |
| *** dguitarbite has joined #openstack-nova | 10:29 | |
| *** mgagne has joined #openstack-nova | 10:29 | |
| *** apmelton has joined #openstack-nova | 10:29 | |
| *** toan has joined #openstack-nova | 10:29 | |
| *** swamireddy has joined #openstack-nova | 10:29 | |
| *** wendar_ has joined #openstack-nova | 10:29 | |
| *** hakimo has joined #openstack-nova | 10:29 | |
| *** hanzhang has joined #openstack-nova | 10:29 | |
| *** adam_g has joined #openstack-nova | 10:29 | |
| *** aix has joined #openstack-nova | 10:29 | |
| *** Kennan has joined #openstack-nova | 10:29 | |
| *** jgriffith has joined #openstack-nova | 10:29 | |
| *** ccarmack has joined #openstack-nova | 10:29 | |
| *** jamielennox has joined #openstack-nova | 10:29 | |
| *** vishwanathj has joined #openstack-nova | 10:29 | |
| *** xgsa has joined #openstack-nova | 10:29 | |
| *** stefanb has joined #openstack-nova | 10:29 | |
| *** sabari has joined #openstack-nova | 10:29 | |
| *** sayali has joined #openstack-nova | 10:29 | |
| *** flaper87 has joined #openstack-nova | 10:29 | |
| *** nikhil_k has joined #openstack-nova | 10:29 | |
| *** jogo has joined #openstack-nova | 10:29 | |
| *** paboldin has joined #openstack-nova | 10:29 | |
| *** lxsli has joined #openstack-nova | 10:29 | |
| *** tcammann has joined #openstack-nova | 10:29 | |
| *** gilliard has joined #openstack-nova | 10:29 | |
| *** tlbr has joined #openstack-nova | 10:29 | |
| bauwser | BobBall-AWOL: quick q, is http://zuul.openstack.xenproject.org/ showing the queue for xen or xenserver ? | 10:30 |
| *** flaper87 has quit IRC | 10:30 | |
| *** flaper87 has joined #openstack-nova | 10:30 | |
| *** haomaiwa_ has quit IRC | 10:30 | |
| *** phschwartz has joined #openstack-nova | 10:30 | |
| bauwser | ah, netsplit | 10:30 |
| bauwser | okay, asking again the quiestion, since we had a netsplt | 10:30 |
| bauwser | BobBall-AWOL: quick q, is http://zuul.openstack.xenproject.org/ showing the queue for xen or xenserver ? | 10:30 |
| *** dguitarbite has quit IRC | 10:30 | |
| *** mgagne has quit IRC | 10:30 | |
| *** apmelton has quit IRC | 10:30 | |
| *** toan has quit IRC | 10:30 | |
| *** swamireddy has quit IRC | 10:30 | |
| *** wendar_ has quit IRC | 10:30 | |
| *** hakimo has quit IRC | 10:30 | |
| *** hanzhang has quit IRC | 10:30 | |
| *** adam_g has quit IRC | 10:30 | |
| *** aix has quit IRC | 10:30 | |
| *** Kennan has quit IRC | 10:30 | |
| *** jgriffith has quit IRC | 10:30 | |
| *** ccarmack has quit IRC | 10:30 | |
| *** jamielennox has quit IRC | 10:30 | |
| *** vishwanathj has quit IRC | 10:30 | |
| *** xgsa has quit IRC | 10:30 | |
| *** stefanb has quit IRC | 10:30 | |
| *** sabari has quit IRC | 10:30 | |
| *** sayali has quit IRC | 10:30 | |
| *** flaper87 has quit IRC | 10:30 | |
| *** nikhil_k has quit IRC | 10:30 | |
| *** jogo has quit IRC | 10:30 | |
| *** paboldin has quit IRC | 10:30 | |
| *** lxsli has quit IRC | 10:30 | |
| *** tcammann has quit IRC | 10:30 | |
| *** gilliard has quit IRC | 10:30 | |
| *** tlbr has quit IRC | 10:30 | |
| *** yixing1 has joined #openstack-nova | 10:31 | |
| *** dguitarbite has joined #openstack-nova | 10:32 | |
| *** mgagne has joined #openstack-nova | 10:32 | |
| *** apmelton has joined #openstack-nova | 10:32 | |
| *** toan has joined #openstack-nova | 10:32 | |
| *** swamireddy has joined #openstack-nova | 10:32 | |
| *** wendar_ has joined #openstack-nova | 10:32 | |
| *** hakimo has joined #openstack-nova | 10:32 | |
| *** hanzhang has joined #openstack-nova | 10:32 | |
| *** adam_g has joined #openstack-nova | 10:32 | |
| *** aix has joined #openstack-nova | 10:32 | |
| *** Kennan has joined #openstack-nova | 10:32 | |
| *** jgriffith has joined #openstack-nova | 10:32 | |
| *** ccarmack has joined #openstack-nova | 10:32 | |
| *** jamielennox has joined #openstack-nova | 10:32 | |
| *** vishwanathj has joined #openstack-nova | 10:32 | |
| *** xgsa has joined #openstack-nova | 10:32 | |
| *** stefanb has joined #openstack-nova | 10:32 | |
| *** sabari has joined #openstack-nova | 10:32 | |
| *** sayali has joined #openstack-nova | 10:32 | |
| *** flaper87 has joined #openstack-nova | 10:32 | |
| *** nikhil_k has joined #openstack-nova | 10:32 | |
| *** jogo has joined #openstack-nova | 10:32 | |
| *** paboldin has joined #openstack-nova | 10:32 | |
| *** lxsli has joined #openstack-nova | 10:32 | |
| *** tcammann has joined #openstack-nova | 10:32 | |
| *** gilliard has joined #openstack-nova | 10:32 | |
| *** tlbr has joined #openstack-nova | 10:32 | |
| openstackgerrit | Noel Nelson Dsouza proposed openstack/nova: Delete flavor should raise error, if it is inuse https://review.openstack.org/217674 | 10:34 |
| *** shoutm has joined #openstack-nova | 10:34 | |
| *** VikasC has quit IRC | 10:34 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add checks for volume status when rebuilding https://review.openstack.org/217567 | 10:36 |
| *** yixing1 has quit IRC | 10:37 | |
| *** atuvenie has quit IRC | 10:37 | |
| *** zhenguo has quit IRC | 10:40 | |
| johnthetubaguy | bauwser: I am happy with that allocation ratio stuff | 10:41 |
| bauwser | johnthetubaguy: cool ! | 10:41 |
| johnthetubaguy | bauwser: thank you for fixing that up, feels must better now | 10:42 |
| bauwser | johnthetubaguy: as I said, I think it's a nice feature for the operatoes | 10:42 |
| johnthetubaguy | not quite as messy as I had feared, either | 10:42 |
| bauwser | johnthetubaguy: because I had many customers and internal bugs related to that | 10:42 |
| johnthetubaguy | bauwser: oh, interesting, thats good data | 10:43 |
| bauwser | johnthetubaguy: we could also change how live-migration is checking the destination because it's broken now because of the allocation | 10:43 |
| bauwser | johnthetubaguy: remember that one https://review.openstack.org/#/c/180151/ | 10:43 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: db: Add the migration_context to the instance_extra table https://review.openstack.org/216731 | 10:43 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: objects: Hook migration object into Instance https://review.openstack.org/216733 | 10:43 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: objects: Add MigrationContext object https://review.openstack.org/216732 | 10:43 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: tests: Make test_claims use Instance object https://review.openstack.org/218201 | 10:43 |
| bauwser | johnthetubaguy: it was a workaround and I didn't provided a new change because I think the allocation ratio BP is better | 10:43 |
| ndipanov | johnthetubaguy, bauwser ^ the DB stuff taken out of the other fixes | 10:44 |
| bauwser | johnthetubaguy: because we could definitely fix _check_destination_has_enough_memory by checking the compute allocation ratios | 10:44 |
| bauwser | ndipanov: oh cool ! | 10:44 |
| johnthetubaguy | ndipanov: I guess that blueprint is all up for review now right? | 10:44 |
| johnthetubaguy | ndipanov: just thinking I am mark it as NeedsCodeReview | 10:44 |
| bauwser | johnthetubaguy: I wonder if we could just say that the DB change is the only feature we need and the others being bugfixes | 10:45 |
| johnthetubaguy | bauwser: I think we did | 10:45 |
| bauwser | okay, then we're good for ndipanov | 10:45 |
| bauwser | ndipanov: I'll review your changes in the next couple of hours - I mean the DB change | 10:46 |
| johnthetubaguy | ndipanov: did you miss the upload on the new patch for this one? https://review.openstack.org/#/c/214713/6 | 10:46 |
| bauwser | but AFAICR, it was a good one | 10:46 |
| ndipanov | johnthetubaguy, weeeelll it's not but it will be today 2 patches missing | 10:47 |
| ndipanov | johnthetubaguy, well I split that into 2 branches | 10:48 |
| ndipanov | will upload one know (as soon as I fix it) | 10:48 |
| *** Marga_ has quit IRC | 10:48 | |
| *** dimsum__ has joined #openstack-nova | 10:53 | |
| *** karimb has quit IRC | 10:55 | |
| *** apuimedo_ has joined #openstack-nova | 10:55 | |
| *** takedakn has joined #openstack-nova | 10:56 | |
| *** bfic has joined #openstack-nova | 10:58 | |
| bauwser | ndipanov: just one comment on https://review.openstack.org/#/c/218201/1 | 11:02 |
| *** bfic has quit IRC | 11:02 | |
| bauwser | ndipanov: basically the problem is that Foo(a=1) works and becomes Foo.a | 11:02 |
| bauwser | ndipanov: it looks like a field, but it's not a field so neither serialized nor coerced | 11:03 |
| *** mnestratov has quit IRC | 11:04 | |
| *** bfic has joined #openstack-nova | 11:05 | |
| *** ildikov has joined #openstack-nova | 11:06 | |
| ndipanov | bauwser, you're confused - instance is a dict there | 11:08 |
| ndipanov | but it should be renamed | 11:09 |
| bauwser | ndipanov: but you're hydrating an object with that dict, right? | 11:09 |
| ndipanov | well no I am passing it to the fake instance_obj function | 11:09 |
| *** smatzek has joined #openstack-nova | 11:09 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Implements detach_interface method https://review.openstack.org/214493 | 11:10 |
| ndipanov | bauwser, which then creades a db instances and calls a _from_db_object | 11:11 |
| ndipanov | which does all the checking | 11:11 |
| ndipanov | bauwser, responded - I think you can remove the -1 | 11:13 |
| bauwser | ndipanov: yeah my bad | 11:14 |
| bauwser | ndipanov: sorry got distracted | 11:14 |
| bauwser | was afk | 11:14 |
| bauwser | yeah, just checked, will remove the -1 | 11:14 |
| *** aix has quit IRC | 11:14 | |
| bauwser | ndipanov: the interface is pretty weak, but _from_db_object() ensures it's good | 11:14 |
| bauwser | ndipanov: hence my confusion | 11:14 |
| ndipanov | bauwser, well if you look at fake_db_obj that gets called - it actually goes and filters out allt he stuff that's not in fields | 11:15 |
| *** bfic has quit IRC | 11:15 | |
| ndipanov | so that thing there is pretty safe by python standards | 11:16 |
| bauwser | ndipanov: agreed | 11:16 |
| bauwser | +1'd | 11:17 |
| * bauwser goes for food | 11:17 | |
| *** thorst1 has joined #openstack-nova | 11:19 | |
| *** bfic has joined #openstack-nova | 11:19 | |
| *** lyanchih has quit IRC | 11:21 | |
| *** takedakn has quit IRC | 11:24 | |
| claudiub | johnthetubaguy: hi. i think there is an issue with how excutils.save_and_reraise_exception works: https://github.com/openstack/nova/blob/master/nova/virt/hyperv/livemigrationops.py#L75 | 11:24 |
| claudiub | johnthetubaguy: it doesn't reraise any exceptions | 11:25 |
| *** takedakn has joined #openstack-nova | 11:26 | |
| claudiub | johnthetubaguy: sample logs: http://paste.openstack.org/show/430812/ | 11:26 |
| *** dprince has joined #openstack-nova | 11:30 | |
| *** marzif has joined #openstack-nova | 11:32 | |
| *** takedakn has quit IRC | 11:35 | |
| *** apuimedo_ has quit IRC | 11:37 | |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: Make rebuild/evacuate use resource claims https://review.openstack.org/214770 | 11:45 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: compute: Make rebuild consider the node that was passed in https://review.openstack.org/214713 | 11:45 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: compute: Make rebuild_instance RPC take additional args https://review.openstack.org/200485 | 11:45 |
| ndipanov | johnthetubaguy, ^ | 11:45 |
| ndipanov | oops there is a bug in that last one now :( renaming FTL | 11:46 |
| johnthetubaguy | claudiub: sorry, getting confused with reading that log, what makes you think its not working? | 11:47 |
| johnthetubaguy | ndipanov: OK, no worries | 11:47 |
| *** lucasagomes is now known as lucas-hungry | 11:48 | |
| *** sudipto has quit IRC | 11:50 | |
| *** kiran-r has quit IRC | 11:51 | |
| johnthetubaguy | claudiub: the lack of "calling live migration post_method" suggest it did re-raise, I think | 11:51 |
| *** apuimedo_ has joined #openstack-nova | 11:51 | |
| johnthetubaguy | claudiub: we are possibly missing a log message in here, to help prove what happened: https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L5004 | 11:53 |
| claudiub | johnthetubaguy: yeah calling live_migration_post_method should happen, but shouldn't the exception be re-raised after it exists the with block? | 12:01 |
| johnthetubaguy | claudiub: it should be, I can't see any evidence either way in those logs, I could be missing something? | 12:02 |
| claudiub | johnthetubaguy: I don't see it being re-raised anywhere either. | 12:02 |
| openstackgerrit | Cedric Brandily proposed openstack/python-novaclient: Allow to reboot multiple servers https://review.openstack.org/218231 | 12:03 |
| johnthetubaguy | claudiub: why would you? there are not logs in that path, hence my message above | 12:03 |
| claudiub | johnthetubaguy: if it's not being re-raised, it's bad, if it's lost, it's still bad | 12:03 |
| *** rushiagr is now known as rushiagr_away | 12:03 | |
| claudiub | johnthetubaguy: i see. Well, a log message would really be helpful | 12:04 |
| johnthetubaguy | claudiub: totally worth adding a log, I suspect you would see it in error notifications and instance faults, and the instance actions, but not in the logs | 12:04 |
| claudiub | johnthetubaguy: yeah... it's hard to track down an exception like this and determine its cause without any logging.. reporting a bug for this and a fix.. | 12:05 |
| *** jwcroppe has quit IRC | 12:06 | |
| johnthetubaguy | claudiub: it is reported, just not in the logs, totally agreed its worth adding a log | 12:06 |
| johnthetubaguy | claudiub: I want to see us agree on the way forward with error handling like this, hence the summit session at the last summit, but we just haven't had anyone with the time to own fixing that setup up yet | 12:08 |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: Switch to oslo.cache lib https://review.openstack.org/203049 | 12:09 |
| claudiub | johnthetubaguy: well, is there a way to at least expose those failures to the user / admin through the api? | 12:13 |
| *** mnestratov has joined #openstack-nova | 12:14 | |
| *** chaochin has quit IRC | 12:14 | |
| *** deepakcs has quit IRC | 12:14 | |
| claudiub | johnthetubaguy: because, at least for live-migration, the instance will just revert to its previous state, it's not immediatly obvious it failed. | 12:15 |
| zigo | Hi there. I'm trying to backport the fix for CVE-2015-3241 to Icehouse, in order to fix nova in Debian jessie. | 12:15 |
| zigo | I've done nearly all, but I'm not sure for the last chunk of the patch at: https://review.openstack.org/#/c/214528/2/nova/virt/libvirt/driver.py | 12:15 |
| zigo | The chunk which adds: | 12:15 |
| zigo | self.job_tracker.terminate_jobs(instance) | 12:16 |
| *** rajesht has quit IRC | 12:16 | |
| zigo | The code is older, and the def delete_instance_files(self, instance): function has a lot less code, so I'm not sure where and how to insert the self.job_tracker.terminate_jobs(instance) code. | 12:16 |
| ndipanov | zigo, so what's the issue with that? | 12:16 |
| *** raildo-afk is now known as raildo | 12:16 | |
| ndipanov | hmmm | 12:16 |
| zigo | Can someone look at the Icehouse version of driver.py and let me know where to call? | 12:16 |
| zigo | The rest of, I believe I figured the backport out myself. | 12:17 |
| zigo | I wouldn't mind someone looking into the patch when I'm done though. | 12:17 |
| *** Kennan has quit IRC | 12:18 | |
| zigo | The function in Icehouse looks like this: http://paste.debian.net/309371/ | 12:18 |
| *** park_ has quit IRC | 12:18 | |
| zigo | I'm tempted to just put it after the if os.path.exists(target): | 12:19 |
| zigo | (the first one) | 12:19 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: compute: Make rebuild_instance RPC take additional args https://review.openstack.org/200485 | 12:20 |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: Switch to oslo.cache lib https://review.openstack.org/203049 | 12:21 |
| *** fawadk has quit IRC | 12:21 | |
| ndipanov | zigo, so I think it's safe to call it anywhere before the rmtree call | 12:22 |
| zigo | Ok. | 12:22 |
| ndipanov | basically it won't do anything if there are no processes started right? | 12:23 |
| zigo | ndipanov: Can you look at the final patch when I'm done with the backport? | 12:23 |
| ndipanov | so it's safe to call it even at the top of the function | 12:23 |
| *** vivekd has quit IRC | 12:23 | |
| ndipanov | zigo, since I need to backport it for our distro too - if you could post a patch that would be very awesome | 12:23 |
| ndipanov | so I'd be happy to review it actually! :) | 12:24 |
| zigo | ndipanov: To where should I post it? | 12:24 |
| ndipanov | zigo, hmmm | 12:24 |
| *** aix has joined #openstack-nova | 12:24 | |
| zigo | ndipanov: Maybe a link to the Debian VCS ? | 12:24 |
| ndipanov | zigo, that works | 12:24 |
| *** deepthi has quit IRC | 12:24 | |
| *** fawadkhaliq has joined #openstack-nova | 12:26 | |
| ndipanov | zigo, ping me here or email me @ my gerrit email | 12:27 |
| ndipanov | when you're done | 12:27 |
| ndipanov | and I'll be happy to review it | 12:27 |
| *** aysyd has joined #openstack-nova | 12:27 | |
| zigo | ndipanov: I'm nearly done. | 12:27 |
| zigo | ndipanov: There's in fact 2 patches, one for oslo-incubator in Nova, and the actual backport. | 12:29 |
| *** Marga_ has joined #openstack-nova | 12:29 | |
| *** rfolco has joined #openstack-nova | 12:29 | |
| *** fawadk has joined #openstack-nova | 12:29 | |
| zigo | ndipanov: The 2 toplevel patches here: http://anonscm.debian.org/cgit/openstack/nova.git?h=debian%2Ficehouse | 12:29 |
| *** fawadkhaliq has quit IRC | 12:30 | |
| zigo | note that it's a patch adding a patch in debian/patches, so you may just want to click on "tree" and download the CVE-2015-3241 patch files. | 12:30 |
| *** salv-orlando has joined #openstack-nova | 12:30 | |
| *** park_ has joined #openstack-nova | 12:30 | |
| ndipanov | zigo, cool will try to look into it today! | 12:31 |
| *** chaochin has joined #openstack-nova | 12:31 | |
| zigo | Thanks. | 12:31 |
| zigo | ndipanov: You can write me back to zigo@debian.org, with 796109@bugs.debian.org and 797192@bugs.debian.org as Cc (note the first one is the actual Debian bug, the 2nd one is the release team bug to ask for an unblock). | 12:32 |
| *** exploreshaifali has quit IRC | 12:32 | |
| zigo | Actually, don't even put zigo@, writing to the bug is enough as it will reach the packaging list. | 12:32 |
| *** bfic has quit IRC | 12:32 | |
| *** salv-orlando has quit IRC | 12:32 | |
| zigo | Hum... | 12:33 |
| *** Kennan has joined #openstack-nova | 12:33 | |
| zigo | Actually, I have to also look at unit test backports, as here, it completely destroyed unit tests. | 12:33 |
| *** beagles has joined #openstack-nova | 12:36 | |
| *** exploreshaifali has joined #openstack-nova | 12:37 | |
| zigo | ndipanov: I can't figure out what broke all unit tests, maybe you can ... | 12:39 |
| zigo | I'll leave things as-is for the moment. | 12:39 |
| *** exploreshaifali has quit IRC | 12:40 | |
| *** exploreshaifali has joined #openstack-nova | 12:41 | |
| *** exploreshaifali has quit IRC | 12:41 | |
| *** lucas-hungry is now known as lucasagomes | 12:44 | |
| *** bfic has joined #openstack-nova | 12:44 | |
| *** haomaiwa_ has joined #openstack-nova | 12:45 | |
| pkholkin | nova cores, please look at two trivial patches https://review.openstack.org/#/c/204082/ https://review.openstack.org/#/c/215658/, it will take a minute, thanks! | 12:45 |
| *** park_ has quit IRC | 12:45 | |
| *** chaochin has quit IRC | 12:45 | |
| *** Kennan has quit IRC | 12:46 | |
| johnthetubaguy | ndipanov: this rebuild stuff where we need the node, thats for when it gets used for evacuate I guess? | 12:46 |
| *** yuntongjin has joined #openstack-nova | 12:47 | |
| johnthetubaguy | ndipanov: just getting confused because rebuild should never move an instance, but evacuate always does | 12:48 |
| *** zul has quit IRC | 12:49 | |
| *** nelsnelson has joined #openstack-nova | 12:50 | |
| *** rushiagr_away is now known as rushiagr | 12:51 | |
| *** fawadk has quit IRC | 12:51 | |
| johnthetubaguy | ndipanov: I am beginning to think evacuate should be a separate compute manager method to rebuild, given (more and more) they are becoming totally different operations? | 12:52 |
| *** zul has joined #openstack-nova | 12:52 | |
| *** takedakn has joined #openstack-nova | 12:52 | |
| *** jcoufal has joined #openstack-nova | 12:53 | |
| *** baoli has joined #openstack-nova | 12:54 | |
| *** mpaolino has joined #openstack-nova | 12:55 | |
| *** mpaolino has quit IRC | 12:55 | |
| *** fawadkhaliq has joined #openstack-nova | 12:55 | |
| *** y_sawai has joined #openstack-nova | 12:55 | |
| openstackgerrit | lalit dagre proposed openstack/nova: Flavor object which support both DB with active migration https://review.openstack.org/213041 | 12:55 |
| openstackgerrit | lalit dagre proposed openstack/nova: Creates flavor* tables in API database https://review.openstack.org/201606 | 12:55 |
| *** takedakn has quit IRC | 12:56 | |
| *** park_ has joined #openstack-nova | 12:57 | |
| *** y_sawai has quit IRC | 12:58 | |
| *** y_sawai has joined #openstack-nova | 12:58 | |
| *** nkrinner has quit IRC | 12:58 | |
| *** takedakn has joined #openstack-nova | 12:59 | |
| *** Kennan has joined #openstack-nova | 13:00 | |
| *** dustins has joined #openstack-nova | 13:00 | |
| *** marzif has quit IRC | 13:00 | |
| *** haomaiwa_ has quit IRC | 13:01 | |
| *** baoli has quit IRC | 13:01 | |
| ndipanov | johnthetubaguy, sorry was out | 13:01 |
| *** baoli has joined #openstack-nova | 13:02 | |
| johnthetubaguy | ndipanov: no worries, just getting my head around the series really | 13:02 |
| *** haomaiwang has joined #openstack-nova | 13:02 | |
| ndipanov | johnthetubaguy, so the node is sent only when evacuate without a host specified | 13:02 |
| johnthetubaguy | ndipanov: yeah, read through that more, makes sense, I just wonder if we would be better having two paths all the way down to the compute manager | 13:03 |
| bauwser | johnthetubaguy: mmm, what's the difference between evacuate and rebuild ? thought it was the same thing, but named differently | 13:03 |
| ndipanov | rebuild will always go to the same host - yes | 13:03 |
| ndipanov | johnthetubaguy, hmmm | 13:03 |
| *** nkrinner has joined #openstack-nova | 13:03 | |
| johnthetubaguy | ndipanov: just seems like every function has the if evacuate do X, etc, wondering if a new path would be cleaner | 13:03 |
| johnthetubaguy | ndipanov: a but like how build and rebuild share code, I guess | 13:04 |
| johnthetubaguy | s/a but/a bit/ | 13:04 |
| *** y_sawai has quit IRC | 13:04 | |
| ndipanov | johnthetubaguy, tbh haven't thought about it let me look | 13:05 |
| ndipanov | but probably does make sense | 13:05 |
| *** mdrabe has joined #openstack-nova | 13:05 | |
| johnthetubaguy | ndipanov: sorry to de rail it, just came to me when I saw the RPC changes | 13:05 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: live-migration: Logs exception if operation failed https://review.openstack.org/217822 | 13:05 |
| *** rlrossit has joined #openstack-nova | 13:06 | |
| ndipanov | johnthetubaguy, np - the most important bits to get in are DB changes in the other branches | 13:07 |
| *** lyanchih has joined #openstack-nova | 13:07 | |
| johnthetubaguy | ndipanov: is that the instance extra migrate bit? | 13:07 |
| *** lyanchih has quit IRC | 13:08 | |
| bauwser | okay, nevermind my silly question about rebuild vs. evacuate | 13:08 |
| bauwser | ndipanov: so you prefer to serialize the object for persisting it, and not a separate table ? | 13:09 |
| bauwser | ndipanov: ie. using instance_extra | 13:09 |
| ndipanov | johnthetubaguy, yes | 13:09 |
| johnthetubaguy | ndipanov: cool, looking through those now | 13:10 |
| *** bfic has quit IRC | 13:10 | |
| ndipanov | bauwser, yeah it was discussed at length there is no reason for a separate table this is never queried only accessed from the instance | 13:10 |
| *** aix has quit IRC | 13:10 | |
| bauwser | ndipanov: okay good to know | 13:10 |
| bauwser | ndipanov: now that we have an object for accessing that, that's fine | 13:11 |
| BobBall-AWOL | bauwser: No. That's for libvirt+xen. XenServer is not using zuul unfortunately (something we hope to change in the future) | 13:13 |
| *** kylek3h has joined #openstack-nova | 13:13 | |
| BobBall-AWOL | bauwser: I think your rechecks should have completed and passed? | 13:13 |
| *** kbyrne has quit IRC | 13:13 | |
| bauwser | BobBall-AWOL: I need to verify | 13:13 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/216968 | 13:16 |
| *** eglynn has quit IRC | 13:16 | |
| *** jwcroppe has joined #openstack-nova | 13:16 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Implements attach_interface and detach_interface method https://review.openstack.org/214493 | 13:17 |
| *** annegentle has joined #openstack-nova | 13:17 | |
| *** kbyrne has joined #openstack-nova | 13:19 | |
| *** kbyrne has quit IRC | 13:20 | |
| gibi | dansmith, sdague: I fixed your comments in https://review.openstack.org/#/c/147514/ could you please check it? | 13:21 |
| *** kbyrne has joined #openstack-nova | 13:21 | |
| openstackgerrit | Feodor Tersin proposed openstack/nova: DNM Adding logging around bdm actions for cells https://review.openstack.org/217829 | 13:23 |
| *** leopoldj has quit IRC | 13:24 | |
| *** nelsnelson has quit IRC | 13:27 | |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: Switch to oslo.cache lib https://review.openstack.org/203049 | 13:28 |
| *** links has quit IRC | 13:29 | |
| *** aix has joined #openstack-nova | 13:29 | |
| *** VW_ has joined #openstack-nova | 13:30 | |
| *** IanGovett has quit IRC | 13:30 | |
| *** IanGovett has joined #openstack-nova | 13:31 | |
| *** amotoki has joined #openstack-nova | 13:32 | |
| *** browne has joined #openstack-nova | 13:33 | |
| *** ferest has joined #openstack-nova | 13:36 | |
| *** mriedem has joined #openstack-nova | 13:37 | |
| *** salv-orlando has joined #openstack-nova | 13:37 | |
| *** thangp has joined #openstack-nova | 13:38 | |
| bauwser | alaski: morning | 13:42 |
| *** park__ has joined #openstack-nova | 13:42 | |
| *** dansmith is now known as superdan | 13:42 | |
| bauwser | alaski: I missed your convo with melwitt on the cells job issues | 13:42 |
| bauwser | alaski: since I see the cells job quite bad these days, I wonder how we could be helping | 13:43 |
| mriedem | bauwser: wasn't that just this https://bugs.launchpad.net/tempest/+bug/1489581 ? | 13:43 |
| openstack | Launchpad bug 1489581 in tempest "test_create_ebs_image_and_check_boot is race failing" [High,In progress] - Assigned to Andrey Pavlov (apavlov-e) | 13:43 |
| bauwser | mriedem: yeah | 13:43 |
| bauwser | mriedem: e-r is not there ? | 13:44 |
| mriedem | there was a patch approved for that yeasterday | 13:44 |
| mriedem | bauwser: not yet | 13:44 |
| mriedem | https://review.openstack.org/#/c/217868/ | 13:44 |
| openstackgerrit | Simona Iuliana Toader proposed openstack/nova: Hyper-V: Adds Hyper-V UEFI Secure Boot https://review.openstack.org/209581 | 13:44 |
| bauwser | mriedem: nice | 13:44 |
| mriedem | let me see if i can find a better logstash query now | 13:44 |
| *** Kennan2 has joined #openstack-nova | 13:44 | |
| *** Kennan has quit IRC | 13:44 | |
| bauwser | mriedem: I had one, lemme find it | 13:45 |
| mriedem | well we had one for the cells job specifically | 13:45 |
| mriedem | message:"Cannot delete snapshot" AND message:"snapshot is busy" AND tags:"screen-c-vol.txt" | 13:45 |
| mriedem | ^ should work for the non-cells jobs | 13:45 |
| ftersin | mriedem: iianm we have two different bugs | 13:45 |
| *** park_ has quit IRC | 13:45 | |
| sdague | gibi: thanks | 13:46 |
| mriedem | maybe, ^ only hits on the ceph job | 13:46 |
| ftersin | mriedem: the first one is "InvalidInput: Invalid input received: Invalid snapshot: Originating snapshot status must be one of'available" | 13:46 |
| *** bnemec has joined #openstack-nova | 13:46 | |
| ftersin | mriedem: got it | 13:46 |
| mriedem | but, i also don't trust logstash so much right now | 13:46 |
| sdague | mriedem: so is the ebs test just crappy? | 13:46 |
| mriedem | sdague: it was just missing a wait for status check | 13:47 |
| *** mmedvede has quit IRC | 13:47 | |
| mriedem | sdague: https://review.openstack.org/#/c/217804/ was the fix | 13:47 |
| mriedem | ftersin: message:"InvalidInput: Invalid input received: Invalid snapshot: Originating snapshot status must be one of 'available" AND tags:"screen-n-cpu.txt" | 13:48 |
| mriedem | does show hits too | 13:48 |
| mriedem | it's odd that this shows different errors in different places depending on the job | 13:48 |
| superdan | would really appreciate a +W on this today: https://review.openstack.org/217239 | 13:48 |
| mriedem | sometimes cells logs, sometimes n-cpu, sometimes c-vol | 13:48 |
| *** edmondsw has joined #openstack-nova | 13:48 | |
| superdan | we really need to get it into L and soak time would be good | 13:48 |
| *** eglynn has joined #openstack-nova | 13:48 | |
| * mriedem adds rlrossit | 13:49 | |
| mriedem | nova-objects-core | 13:49 |
| superdan | heh | 13:49 |
| gibi | sdague: thanks for your review :) | 13:49 |
| sdague | mriedem: are there other fixing changes that should be chased? | 13:49 |
| *** alaski is now known as lascii | 13:49 | |
| mriedem | sdague: not that i'm aware of | 13:49 |
| sdague | ok | 13:49 |
| *** dimsum__ has quit IRC | 13:49 | |
| mriedem | i added nova to the bug in LP though since it seems nova could be doing some status checking of it's own when given the snapshot | 13:50 |
| sdague | I'll hound infra folks on a promote for that one | 13:50 |
| *** mmedvede has joined #openstack-nova | 13:50 | |
| *** dimsum__ has joined #openstack-nova | 13:50 | |
| *** ctrath has joined #openstack-nova | 13:50 | |
| *** snikitin has quit IRC | 13:50 | |
| *** bfic has joined #openstack-nova | 13:51 | |
| *** rushiagr is now known as rushiagr_away | 13:51 | |
| bauwser | superdan: oh nice change ! | 13:51 |
| bauwser | superdan: so the client is sending all the versions he knows so the manager backports the object to the specific target by considering the client versions? | 13:52 |
| bauwser | nice | 13:52 |
| superdan | yeah | 13:52 |
| superdan | it's going to make things a lot nicer around here | 13:52 |
| bauwser | yup | 13:53 |
| *** eharney has joined #openstack-nova | 13:53 | |
| superdan | no more bumping BDM every time instance changes, etc | 13:53 |
| *** bfic has quit IRC | 13:53 | |
| bauwser | I guess the foundery is all in o.vo ? | 13:53 |
| mriedem | sdague: bauwser: ftersin: i think i have the logstash query to end all logstash queries for this ebs test bug now | 13:53 |
| superdan | bauwser: yeah, it is now | 13:53 |
| bauwser | superdan: cool | 13:53 |
| bauwser | (that's what I can understand, because your change is Jenkins compatible :) ) | 13:53 |
| *** rlrossit is now known as rlrooster | 13:53 | |
| sdague | mriedem: is test_ids there for anything other than information? | 13:54 |
| mriedem | sdague: logstash was failing me yesterday, long story | 13:54 |
| mriedem | anyway, pushing new query now | 13:54 |
| sdague | oh, well I approved the old one | 13:55 |
| *** snikitin has joined #openstack-nova | 13:55 | |
| mriedem | gah | 13:55 |
| mriedem | ok | 13:55 |
| mriedem | updating then | 13:55 |
| ftersin | mriedem: something like this: http://logstash.openstack.org/#eyJzZWFyY2giOiJcIkZpbGUgXFxcInRlbXBlc3Qvc2NlbmFyaW8vdGVzdF92b2x1bWVfYm9vdF9wYXR0ZXJuLnB5XFxcIiwgbGluZSAxOTQsIGluIHRlc3RfY3JlYXRlX2Vic19pbWFnZV9hbmRfY2hlY2tfYm9vdFwiIiwiZmllbGRzIjpbXSwib2Zmc2V0IjowLCJ0aW1lZnJhbWUiOiIxNzI4MDAiLCJncmFwaG1vZGUiOiJjb3VudCIsInRpbWUiOnsidXNlcl9pbnRlcnZhbCI6MH0sInN0YW1wIjoxNDQwNzcwMDkwMjM3fQ== | 13:55 |
| johnthetubaguy | superdan: for this manifest change, should we have a follow on one that drops the old backport method, or do we still use that somewhere? | 13:55 |
| mriedem | ftersin: nope | 13:55 |
| rlrooster | johnthetubaguy: you need to keep that around to maintain the API | 13:56 |
| superdan | johnthetubaguy: we can't drop it until M at least, and then it requires a conductor rpc bump to get rid of it | 13:56 |
| bauwser | johnthetubaguy: we need to keep a compat API | 13:56 |
| *** Samolo_ has joined #openstack-nova | 13:56 | |
| bauwser | oh man | 13:56 |
| Samolo_ | Hi all ! , is there any flag to tell nova or another services to restart an instance after beeing move by a host evacuate please ? | 13:56 |
| johnthetubaguy | superdan: I was just thinking about the RPC API side, not the manager method, thats sure has to stick around | 13:56 |
| rlrooster | superdan: can we even drop it? o.vo has to keep it around forever right? | 13:57 |
| *** edleafe is now known as figleaf | 13:57 | |
| *** fawadk has joined #openstack-nova | 13:57 | |
| superdan | rlrooster: o.vo does yeah, but we don't have to implement it.. most projects don't | 13:57 |
| rlrooster | leaving an abstract method as NotImplemented? | 13:57 |
| superdan | johnthetubaguy: yeah, we can drop the client side of it, but that should be a separate patch I think, like all our conductor deprecation patches | 13:57 |
| * rlrooster gets scared | 13:57 | |
| ftersin | mriedem: i think it shows all failures of this test (with doubled bdm and not available image) | 13:57 |
| superdan | rlrooster: it's not abstract, it's an implement-it-if-you-want-it thing | 13:58 |
| superdan | nova is literally the only project that uses it :) | 13:58 |
| superdan | everyone else just leaves it unimplemented | 13:58 |
| rlrooster | oh yep that one is just a pass | 13:58 |
| rlrooster | object_backport_versions is the NotImplemented one | 13:59 |
| johnthetubaguy | superdan: yeah, follow on patch makes sense, just checking to make sure I understood what was happening really | 13:59 |
| superdan | rlrooster: and that was so we can tell if they've moved from the old to the new one | 13:59 |
| openstackgerrit | jichenjc proposed openstack/nova: Catch invalid id input in service_delete https://review.openstack.org/198753 | 13:59 |
| superdan | johnthetubaguy: yep | 13:59 |
| mriedem | ftersin: https://review.openstack.org/#/c/218276/ | 13:59 |
| mriedem | ftersin: we prefer to not write queries against things in the console or tempest logs when we can use errors from the service screen logs | 13:59 |
| *** Kennan2 has quit IRC | 14:00 | |
| mriedem | http://goo.gl/AGU0rG | 14:00 |
| *** fawadkhaliq has quit IRC | 14:00 | |
| *** dprince has quit IRC | 14:01 | |
| *** haomaiwang has quit IRC | 14:01 | |
| *** park__ has quit IRC | 14:01 | |
| bauwser | ndipanov: superdan: not sure you have time for that, but there is a green light on https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/allocation-ratio-to-resource-tracker,n,z | 14:01 |
| johnthetubaguy | superdan: do we need to get this bumped (via global) so we can use this method? https://github.com/openstack/nova/blob/master/requirements.txt#L52 | 14:01 |
| Samolo_ | Hi all ! , is there any flag to tell nova or another services to restart an instance after beeing move by a host evacuate please ? | 14:01 |
| *** jwcroppe has quit IRC | 14:01 | |
| Samolo_ | or migrate | 14:01 |
| superdan | johnthetubaguy: rlrooster has it in the pipeline | 14:02 |
| *** haomaiwang has joined #openstack-nova | 14:02 | |
| johnthetubaguy | superdan: sweet | 14:02 |
| rlrooster | johnthetubaguy: https://review.openstack.org/#/c/216848/ | 14:02 |
| mriedem | Samolo_: not that i'm aware of, you could just call the reboot api once your evacuate/migration is done | 14:02 |
| *** kaisers has quit IRC | 14:02 | |
| rlrooster | johnthetubaguy: but we need lifeless to get https://review.openstack.org/#/c/216992/ in before mine will go | 14:03 |
| Samolo_ | mriedem: oh :( sad to ear that | 14:03 |
| Samolo_ | mriedem: thanks | 14:03 |
| mriedem | Samolo_: i guess you'd have to explain the use case | 14:03 |
| mriedem | evacuate rebuilds the instance on another host so i don't know why you'd need to reboot it after that | 14:03 |
| *** lyanchih has joined #openstack-nova | 14:05 | |
| *** andreykurilin__ has quit IRC | 14:05 | |
| johnthetubaguy | ndipanov: I am trying to get straight in my head why we want to create migration_context, rather than just add the topology into the migration object itself: https://review.openstack.org/#/c/216732/3/nova/objects/migration_context.py,cm | 14:06 |
| johnthetubaguy | ndipanov: feel like I am missing a piece of the puzzle | 14:06 |
| Samolo_ | mriedem: thanks, I will read more on the mechanism I think it will be useffull | 14:06 |
| ndipanov | johnthetubaguy, a very detailed discussion on that very issue here https://review.openstack.org/163440 | 14:06 |
| ndipanov | gerrit is of course a horrible place for it since there is no way for people to find it but that's what we have | 14:07 |
| *** Piet has quit IRC | 14:07 | |
| johnthetubaguy | Samolo_: FWIW migrate usually turns of the instance, moves it, then starts it on the other side (as opposed to live migrate) | 14:07 |
| *** exploreshaifali has joined #openstack-nova | 14:08 | |
| ndipanov | johnthetubaguy, plus once you see the code that uses it | 14:08 |
| ndipanov | (which I may post once I fix the tests which is turnign out to be a horrible task) | 14:08 |
| ndipanov | it will make a lot of sense | 14:09 |
| ndipanov | it's almos always used from the instance | 14:09 |
| *** yuntongjin has quit IRC | 14:09 | |
| ndipanov | so I just add Instance.apply_mig_context() | 14:09 |
| ndipanov | instance.revert_mig_context() | 14:09 |
| mriedem | ndipanov: ftersin: would be good if you guys could review this at some point https://review.openstack.org/#/c/215311/ | 14:09 |
| mriedem | it's a dependency for an lxc ci job | 14:10 |
| johnthetubaguy | ndipanov: cool, that extra change is good context | 14:10 |
| ndipanov | mriedem, will try | 14:10 |
| Samolo_ | johnthetubaguy: thanks for the information | 14:10 |
| johnthetubaguy | ndipanov: avoid storing a tun of data for ever is a very good reason, knew I was missing something | 14:11 |
| johnthetubaguy | s/tun/ton/ | 14:11 |
| johnthetubaguy | or something | 14:11 |
| lxsli | tun, ton and tonne are all valid units :) | 14:11 |
| *** yuntongjin has joined #openstack-nova | 14:12 | |
| superdan | johnthetubaguy: yeah, I'm +1 on the context as opposed to putting it on the migration | 14:12 |
| ndipanov | johnthetubaguy, that too we just null it out when done | 14:13 |
| ndipanov | superdan, yeah it was a good call at the time | 14:13 |
| *** artom has joined #openstack-nova | 14:13 | |
| johnthetubaguy | lxsli: oh year, mash tun makes more sense now | 14:13 |
| *** park__ has joined #openstack-nova | 14:14 | |
| *** Kennan has joined #openstack-nova | 14:14 | |
| johnthetubaguy | superdan: ndipanov: yeah I am seeing the reasoning now, I like it | 14:14 |
| *** Samolo_ has quit IRC | 14:14 | |
| ndipanov | johnthetubaguy, once you see the follow on patches it will make even more sense - having it on the migration would have been way more code | 14:15 |
| *** jichen has quit IRC | 14:15 | |
| ndipanov | (some time this century I'm sure) | 14:15 |
| johnthetubaguy | ndipanov: cool | 14:16 |
| lascii | bauwser: alright, I'm around now. (long morning with car troubles for the wife) | 14:17 |
| lascii | bauwser: for the cells thing my plan is to dig through the logs on https://review.openstack.org/#/c/217829/ | 14:17 |
| lascii | and see what's going on there. beyond that I'm not sure at this point | 14:18 |
| *** tpeoples has joined #openstack-nova | 14:19 | |
| mriedem | lascii: pretty sure the cells thing is just https://bugs.launchpad.net/tempest/+bug/1489581 | 14:19 |
| openstack | Launchpad bug 1489581 in tempest "test_create_ebs_image_and_check_boot is race failing" [High,In progress] - Assigned to Andrey Pavlov (apavlov-e) | 14:19 |
| lascii | mriedem: okay. I'll see if I can confirm that | 14:20 |
| *** yuntongjin has quit IRC | 14:21 | |
| openstackgerrit | chandan kumar proposed openstack/nova: Removed unused dependency: discover https://review.openstack.org/218286 | 14:21 |
| *** zhenguo has joined #openstack-nova | 14:22 | |
| *** aix has quit IRC | 14:25 | |
| *** agireud has joined #openstack-nova | 14:26 | |
| *** shuquan has joined #openstack-nova | 14:27 | |
| *** yamahata has joined #openstack-nova | 14:28 | |
| *** burgerk has joined #openstack-nova | 14:32 | |
| *** agireud has quit IRC | 14:32 | |
| *** stefan_amann has quit IRC | 14:32 | |
| lxsli | bauwser: hiya, nova.scheduler.utils.set_vm_state_and_notify gets instance_properties from the request_spec and puts that into a payload dict | 14:34 |
| lxsli | bauwser: instance_properties seems to have been disassembled onto the RequestSpec object, do you think I should reconstitute it in set_vm_state_and_notify or what should I do? | 14:36 |
| *** burgerk_ has joined #openstack-nova | 14:36 | |
| *** dprince has joined #openstack-nova | 14:37 | |
| *** aix has joined #openstack-nova | 14:38 | |
| openstackgerrit | Ed Leafe proposed openstack/nova: Adds unit tests to test_common.py https://review.openstack.org/218307 | 14:38 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Rename tests so that they are run https://review.openstack.org/218308 | 14:38 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Replace os.path.join() for URLs https://review.openstack.org/218309 | 14:38 |
| *** exploreshaifali has quit IRC | 14:39 | |
| *** burgerk has quit IRC | 14:39 | |
| *** takedakn has quit IRC | 14:40 | |
| bauwser | lascii: lxsli: sorry was afk | 14:42 |
| lxsli | wb | 14:42 |
| bauwser | lascii: eh, no worries, mriedem told me the main bug | 14:42 |
| bauwser | lascii: https://bugs.launchpad.net/tempest/+bug/1489581 | 14:42 |
| openstack | Launchpad bug 1489581 in tempest "test_create_ebs_image_and_check_boot is race failing" [High,In progress] - Assigned to Andrey Pavlov (apavlov-e) | 14:42 |
| bauwser | lascii: so let's see what happens in the next days because there is a bugfix | 14:42 |
| lascii | okay. I'm working through it to see if matches the cells failure. but hopefully so | 14:43 |
| bauwser | lascii: yup it does | 14:43 |
| bauwser | lascii: tho we have some trace for the region cells, you know, but that's not the problem - | 14:44 |
| bauwser | lascii: I mean, the Conflict trace | 14:44 |
| *** mtanino has joined #openstack-nova | 14:44 | |
| bauwser | lascii: I know it's not a bug, just wondering how we can catch it to not provide it | 14:44 |
| bauwser | lascii: because it could be confusing | 14:45 |
| *** ferest has quit IRC | 14:45 | |
| lascii | we could catch it and log a warning instead | 14:45 |
| *** kbyrne has quit IRC | 14:45 | |
| bauwser | lascii: I mean http://logs.openstack.org/86/188486/30/check/gate-tempest-dsvm-cells/78a34c9/logs/screen-n-cell-region.txt.gz?level=TRACE#_2015-08-27_18_18_43_485 | 14:46 |
| bauwser | lascii: okay, since I'm pretty done with my changes, I'll do that | 14:46 |
| bauwser | lascii: like last cycle, I have more time for helping the cells by FF | 14:47 |
| bauwser | lascii: last point, I know you asked me a few stuff in https://review.openstack.org/#/c/187197/ | 14:48 |
| bauwser | lascii: since I'm pretty interesting in getting your opinions for that BP, could you please review it once you have time ? | 14:48 |
| *** aix has quit IRC | 14:48 | |
| *** doude has quit IRC | 14:48 | |
| openstackgerrit | Chuck Carmack proposed openstack/nova: Don't trace on InstanceInfoCacheNotFound when refreshing network info_cache https://review.openstack.org/217840 | 14:48 |
| bauwser | lxsli: I haven't forgotten you | 14:48 |
| bauwser | lxsli: so, the thing is, we decided to split out the instance object into separate fields | 14:49 |
| *** doude has joined #openstack-nova | 14:49 | |
| *** shuquan has quit IRC | 14:49 | |
| bauwser | lxsli: we also plan to do that for the Flavor object, but we haven't yet had time because there is a filter which uses some Flavor fields directly | 14:50 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: DO NOT MERGE: replace global CONF access by object https://review.openstack.org/218319 | 14:50 |
| bauwser | lxsli: but consider that the goal of the ReqSpec object is to not have nested objects, but rather separate fields that can by hydrated from other objects | 14:50 |
| markus_z | meh... sorry... premature push... | 14:50 |
| bauwser | markus_z: http://docs.openstack.org/developer/oslo.config/faq.html#why-does-oslo-config-have-a-conf-object-global-objects-suck | 14:51 |
| lxsli | bauwser: not sure what your conclusion is? I should make up a dict that looks like the old instance_properties? | 14:52 |
| markus_z | bauwser: yeah, I'm still young and dumb enough to try it :) | 14:52 |
| markus_z | bauwser: http://lists.openstack.org/pipermail/openstack-dev/2015-July/070306.html | 14:52 |
| bauwser | lxsli: that's a freeform payload | 14:52 |
| bauwser | lxsli: so I proposed to send an object rather | 14:53 |
| lxsli | bauwser: we don't need to worry about API stability? | 14:53 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: DO NOT MERGE: replace global CONF access by object https://review.openstack.org/218319 | 14:53 |
| bauwser | lxsli: which API? notifications? lemme laugh | 14:53 |
| lxsli | ok well glad we cleared that up :D | 14:53 |
| bauwser | lxsli: I mean | 14:54 |
| lxsli | so I'll just send request_spec instead of instance_properties then | 14:54 |
| bauwser | lxsli: I remember some NEED_ACTION item about notifications in YVE | 14:54 |
| bauwser | YVR | 14:54 |
| bauwser | lxsli: I guess I'm probably really cowboy when saying I'm sending an object to the wire | 14:55 |
| bauwser | lxsli: but I think it's the best thing to do, because we serialize/unserialize properly | 14:55 |
| bauwser | lxsli: but I'm waiting comments on the change I made where I sent an object - so I guess it's debatable | 14:56 |
| *** meghal has joined #openstack-nova | 14:56 | |
| lxsli | bauwser: OK, thanks. This is my first big patch so I'm happy to just follow advice and see what comments turn up | 14:56 |
| bauwser | lxsli: zactly what I did | 14:56 |
| *** rushiagr_away is now known as rushiagr | 14:57 | |
| *** shuquan has joined #openstack-nova | 14:57 | |
| *** kbyrne has joined #openstack-nova | 14:57 | |
| lascii | ndipanov: I think we're seeing a race in bdm_update_or_create being called twice in quick succession. There are no unique constraints for bdms, should there be? | 14:57 |
| *** meghal1 has joined #openstack-nova | 14:58 | |
| bauwser | lxsli: well, that's not true for all I did I remember now | 14:58 |
| ndipanov | lascii, where does this happen? API? | 14:58 |
| bauwser | lxsli: I also used the _to_legacy_spec_dict() | 14:58 |
| bauwser | ndipanov: I think lxsli wanted to mention the payload | 14:58 |
| *** mdrabe has quit IRC | 14:59 | |
| ndipanov | there are none in the DB lascii but there is a soft constraint on not having same device names for an instnace | 14:59 |
| lascii | ndipanov: cells. so at the api level, yes. but it's quick updates on the bdm in the compute that's causing it to be called | 14:59 |
| ndipanov | lascii, and what's the race? | 14:59 |
| *** aix has joined #openstack-nova | 15:00 | |
| lascii | block_device_mapping_update_or_create db method called twice in quick succession causes each to create an entry | 15:00 |
| *** meghal has quit IRC | 15:00 | |
| bauwser | ndipanov: http://logs.openstack.org/86/188486/30/check/gate-tempest-dsvm-cells/78a34c9/console.html#_2015-08-27_18_34_29_976 | 15:00 |
| lascii | http://logs.openstack.org/29/217829/3/check/gate-tempest-dsvm-cells/dd50b22/logs/screen-n-cell-region.txt.gz#_2015-08-28_14_33_47_605 and a few lines above | 15:00 |
| *** haomaiwang has quit IRC | 15:01 | |
| bauwser | lxsli: https://review.openstack.org/#/c/188487/30/nova/scheduler/filter_scheduler.py,cm L72 | 15:01 |
| ndipanov | hmmm interesting | 15:01 |
| *** pkholkin has quit IRC | 15:01 | |
| *** haomaiwang has joined #openstack-nova | 15:02 | |
| lxsli | bauwser: as it happens the request_spec was already in the notification payload so I just deleted the instance_properties from the payload | 15:02 |
| lxsli | test passes, wooooo | 15:02 |
| bauwser | lxsli: I don't get all the context but that's cool | 15:03 |
| *** zzzeek has joined #openstack-nova | 15:03 | |
| *** mtanino has quit IRC | 15:03 | |
| *** snikitin has quit IRC | 15:03 | |
| ndipanov | so lascii correct me if I'm wrong | 15:03 |
| ndipanov | a boot request would _not_ create BDMs in the api cell | 15:04 |
| *** pkholkin has joined #openstack-nova | 15:04 | |
| ndipanov | then it would only create them in the cpu cell and propagate that upwards | 15:04 |
| *** snikitin has joined #openstack-nova | 15:05 | |
| *** dave-mccowan has quit IRC | 15:06 | |
| lascii | correct | 15:06 |
| ndipanov | hmmm | 15:06 |
| lascii | it's the propagation upwards that's racing afaict | 15:06 |
| *** hemnafk is now known as hemna | 15:06 | |
| ndipanov | lascii, can you give me the link to the patch - I think you added some logging | 15:07 |
| *** tonytan4ever has joined #openstack-nova | 15:07 | |
| lascii | https://review.openstack.org/#/c/217829 | 15:07 |
| *** whenry_ has joined #openstack-nova | 15:07 | |
| ndipanov | lascii, but looking at the DB code it should not create 2 records even if it does race... | 15:08 |
| ndipanov | hmmm I think that if you remove the create_check here https://review.openstack.org/#/c/217829/3/nova/cells/messaging.py | 15:09 |
| bauwser | lascii: so it creates 2 BDM records for the same bdm ? | 15:10 |
| ndipanov | it would probably solve the issue | 15:10 |
| bauwser | lascii: one in the child cell, one in the api cell ? | 15:10 |
| ndipanov | since update_or_create will not allow for 2 records | 15:10 |
| ndipanov | however I am unclear as to why it races | 15:10 |
| *** kylek3h has quit IRC | 15:10 | |
| ndipanov | lascii, I am not sure why is there a check like that | 15:11 |
| lascii | on a call with johnthetubaguy so a bit slow atm | 15:11 |
| ndipanov | say hi | 15:11 |
| ndipanov | :) | 15:11 |
| johnthetubaguy | hi | 15:11 |
| ndipanov | hi | 15:12 |
| *** armax has joined #openstack-nova | 15:12 | |
| *** zhenguo has quit IRC | 15:14 | |
| *** boris-42 has joined #openstack-nova | 15:14 | |
| *** mdrabe has joined #openstack-nova | 15:15 | |
| *** Nikolay_St has quit IRC | 15:16 | |
| *** doude has quit IRC | 15:16 | |
| *** doude_ has joined #openstack-nova | 15:16 | |
| *** gabriel-bezerra is now known as gabrielb | 15:19 | |
| *** edmondsw has quit IRC | 15:19 | |
| *** dave-mccowan has joined #openstack-nova | 15:19 | |
| *** gabrielb is now known as gabriel-bezerra | 15:20 | |
| *** edmondsw has joined #openstack-nova | 15:20 | |
| *** gabriel-bezerra is now known as gbezerra | 15:22 | |
| *** gbezerra is now known as gabezerra | 15:22 | |
| *** gabezerra is now known as gbezerra_ | 15:23 | |
| *** gbezerra_ is now known as gabrielb | 15:23 | |
| *** jaypipes has joined #openstack-nova | 15:23 | |
| *** gabrielb is now known as gabriel-bezerra | 15:23 | |
| *** jaypipes is now known as leakypipes | 15:24 | |
| *** apuimedo_ has quit IRC | 15:25 | |
| *** dguitarbite has quit IRC | 15:26 | |
| leakypipes | mornin all | 15:26 |
| superdan | leakypipes: are you still in PDT? | 15:27 |
| openstackgerrit | Andrey Kurilin proposed openstack/python-novaclient: Launch functional tests serially https://review.openstack.org/217768 | 15:28 |
| *** shuquan has quit IRC | 15:28 | |
| bauwser | leakypipes: mourninge | 15:30 |
| *** ferest has joined #openstack-nova | 15:30 | |
| bauwser | leakypipes: so I rebased your pci_cleanup branch | 15:31 |
| figleaf | leakypipes: good UGT morning | 15:31 |
| bauwser | leakypipes: and that faced me off that you're wedging ndipanov's work :p | 15:31 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Remove 'v3' from extension code https://review.openstack.org/214311 | 15:31 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Remove v3 references in unit test 'contrib' https://review.openstack.org/212851 | 15:31 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Remove more 'v3' references from the code https://review.openstack.org/214290 | 15:31 |
| *** fawadk has quit IRC | 15:32 | |
| leakypipes | superdan: sitting in SJC waiting to get on a plane back home... so yes, for an hour or so ;) | 15:32 |
| bauwser | leakypipes: I mean https://review.openstack.org/#/q/status:merged+project:openstack/nova+branch:master+topic:bp/migration-fix-resource-tracking,n,z | 15:32 |
| superdan | leakypipes: ah | 15:32 |
| *** skylerberg has joined #openstack-nova | 15:33 | |
| leakypipes | bauwser: how am I wedging ndipanov's work? | 15:33 |
| *** tpeoples has quit IRC | 15:33 | |
| bauwser | leakypipes: nah, I just mean that ndipanov is providing new tests for the migration stuff | 15:33 |
| *** neelabh_ has quit IRC | 15:33 | |
| bauwser | leakypipes: and that conflicts with you removing those tests and modifying test_tracker | 15:34 |
| *** electroc_ has joined #openstack-nova | 15:34 | |
| *** snikitin has quit IRC | 15:36 | |
| leakypipes | bauwser: ah, I see... well, I'll rebase if his stuff lands first, no worries. | 15:36 |
| figleaf | johnthetubaguy: can you re-add your +W to https://review.openstack.org/#/c/212851/ ? I pushed a new PS for a dependent patch, and it wiped it out of this one. | 15:36 |
| bauwser | leakypipes: yeah there is no clear problem, just a FYI | 15:36 |
| bauwser | leakypipes: since you're basically moving tests, you're not impacted by FF so we can land ndipanov's patches first | 15:36 |
| *** zzzeek has quit IRC | 15:37 | |
| *** zzzeek has joined #openstack-nova | 15:38 | |
| *** zzzeek has quit IRC | 15:38 | |
| leakypipes | bauwser: yup, makes sense. | 15:38 |
| *** tdurakov has quit IRC | 15:39 | |
| *** zzzeek has joined #openstack-nova | 15:39 | |
| *** tjones has joined #openstack-nova | 15:39 | |
| *** pkholkin has quit IRC | 15:40 | |
| bauwser | leakypipes: before you leave, are we still planning to deliver resource-objects ? | 15:40 |
| bauwser | leakypipes: I can see an easily fixable docs issue on https://review.openstack.org/#/c/128992/27, okay with me proposing another set? | 15:41 |
| openstackgerrit | Brian Elliott proposed openstack/nova: Add migration policy to upgrades devref https://review.openstack.org/218362 | 15:42 |
| *** hammondr has joined #openstack-nova | 15:42 | |
| hammondr | can anyone help with a hypervisor node and rabbitmq problem? | 15:42 |
| superdan | sdague: the microversion thread from this morning.. didn't we *not* add the idea of "latest" on the server side? | 15:42 |
| *** rcarrillocruz has joined #openstack-nova | 15:42 | |
| *** tpeoples has joined #openstack-nova | 15:44 | |
| sdague | superdan: no, it's still in there - https://github.com/openstack/nova/blob/7a1ef0b53874d3a596583aa5ad451ae3c72637f2/nova/api/openstack/wsgi.py#L223 | 15:44 |
| sdague | we just don't advertise it or tell people they should use it | 15:44 |
| superdan | ugh | 15:44 |
| mriedem | i'm planning on updating the docs with a warning on that | 15:45 |
| johnthetubaguy | figleaf: sure, +W | 15:46 |
| lascii | ndipanov: bauwser back | 15:46 |
| figleaf | johnthetubaguy: thx | 15:46 |
| lascii | ndipanov: so for the create check here https://review.openstack.org/#/c/217829/3/nova/cells/messaging.py I have yet to see create be True | 15:46 |
| bauwser | those Rackspace meetings are quite short :D | 15:46 |
| lascii | ndipanov: so I can remove it, but I'm not sure it'll help | 15:46 |
| *** mtanino has joined #openstack-nova | 15:46 | |
| bauwser | lascii: so I'm basically trying to understand the logic | 15:47 |
| lascii | bauwser: heh, it was just johnthetubaguy and I so we could be efficient | 15:47 |
| bauwser | lascii: so we create the BDM on the child cell because it's where it has to be done, but then we also update the parent cell with the bdm information to sync them, right? | 15:47 |
| bauwser | lascii: that's what I can understand | 15:48 |
| ndipanov | lascii, hmmm | 15:48 |
| *** vivekd has joined #openstack-nova | 15:48 | |
| *** tdurakov has joined #openstack-nova | 15:48 | |
| ndipanov | well unless the update_or_create db method is not using transactions properly | 15:48 |
| ndipanov | I don't see how ti could be racy then | 15:48 |
| *** rushiagr is now known as rushiagr_away | 15:51 | |
| mriedem | i'm tempted to poke a mr booth about racy db transactions :) | 15:51 |
| mriedem | and then leave the room | 15:51 |
| ndipanov | mriedem, lol | 15:52 |
| superdan | I like racy content | 15:52 |
| *** ircuser-1 has quit IRC | 15:52 | |
| lascii | This code has been rated R by the ESRB | 15:52 |
| ndipanov | hah | 15:53 |
| lascii | bauwser: yes, we skip creating the bdm in the api cell initially. the execution gets to the child cell which creates the bdm there and syncs it up | 15:53 |
| bauwser | lascii: gotchaz | 15:53 |
| openstackgerrit | Pengtao Huang proposed openstack/nova: Replace "vol" variable by "bdm" https://review.openstack.org/208211 | 15:54 |
| *** vladikr has joined #openstack-nova | 15:54 | |
| *** shoutm has quit IRC | 15:54 | |
| *** ferest has quit IRC | 15:55 | |
| leakypipes | bauwser: yes I was definitely planning for resource-objects. the pci cleanup stuff is kind of a prerequisite, which is why I have been trying to push that. | 15:55 |
| *** shoutm has joined #openstack-nova | 15:55 | |
| *** shoutm has quit IRC | 15:55 | |
| bauwser | leakypipes: okay | 15:55 |
| *** matrohon has quit IRC | 15:56 | |
| bauwser | leakypipes: sorry if you think it was a frank question, I was asking because of FF pretty close | 15:56 |
| leakypipes | bauwser: I need to rebase the resource-objects HEAD onto the pci-cleanup HEAD. will work on it on the plane. | 15:56 |
| bauwser | leakypipes: oh ok that was my question | 15:56 |
| leakypipes | bauwser: no worries, Sylvain, glad you brought it up! | 15:56 |
| *** haomaiwang has quit IRC | 15:58 | |
| lascii | ndipanov: it's possible that if https://github.com/openstack/nova/blob/master/nova/db/sqlalchemy/api.py#L3924 was running in two greenthreads that they could both come up with an empty result right? | 15:58 |
| lascii | and then both would create a new bdm | 15:58 |
| *** haomaiwang has joined #openstack-nova | 15:58 | |
| ndipanov | lascii, how so - I thought that with session.begin() == db transaction | 15:59 |
| superdan | lascii: yes | 15:59 |
| mriedem | how does everyone feel about a db migration to add a unique constraint this close to rc1? https://review.openstack.org/#/c/212908/ | 15:59 |
| superdan | ndipanov: AFAIK, the transaction just freezes the state of the world for that session, which means a query..insert can happen in both cases, and unless there is a where clause on the insert, or a unique constraint, they'll both succeed | 16:00 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Implements attach_interface and detach_interface method https://review.openstack.org/214493 | 16:00 |
| *** haomaiwang has quit IRC | 16:01 | |
| lascii | that's what I was thinking. that's why I'm wondering about adding a uniqe constraint | 16:01 |
| mriedem | phschwartz: do you have access to some perf test systems at rax to run this through and see if it's ok? https://review.openstack.org/#/c/212908/ | 16:01 |
| sdague | mriedem: so is there recovery there for when it would attempt to happen? | 16:01 |
| mriedem | sdague: which thing? | 16:01 |
| sdague | the unique constraint bit | 16:01 |
| sdague | I guess the big concern is what happens with dbs that have raced this way in the past | 16:01 |
| ndipanov | superdan, so we need explicitly where | 16:02 |
| *** haomaiwang has joined #openstack-nova | 16:02 | |
| ndipanov | you are probably right | 16:02 |
| sdague | the migration doesn't seem to account for that, which would be a migration fail | 16:02 |
| mriedem | sdague: yeah | 16:02 |
| phschwartz | mriedem: if your saying you want me to pull it down to run through the CI tests for it on a perf host, I should be able to | 16:02 |
| ndipanov | lascii, so that would help | 16:02 |
| mriedem | phschwartz: yeah | 16:02 |
| phschwartz | mriedem: give me a few, I will do that. | 16:02 |
| mriedem | phschwartz: basically, i proxy requests to you that i used to send to jerdfelt | 16:02 |
| superdan | the unique constraint seems good to me | 16:02 |
| mriedem | superdan: unless the db has duplicates already | 16:02 |
| superdan | (for the ndipanov/lascii thread) | 16:02 |
| ndipanov | superdan, on (instance, device_name) | 16:03 |
| phschwartz | mriedem: not a problem. That is part of why I am here. | 16:03 |
| ndipanov | superdan, on (instance, device_name, deleted) | 16:03 |
| superdan | ndipanov: aye | 16:03 |
| lascii | ndipanov: yeah | 16:03 |
| ndipanov | sigh | 16:03 |
| lascii | ndipanov: unless there's a reason that would be a problem for bdms | 16:03 |
| sdague | mriedem: in general it seems like this would be a reasonable constraint, but addressing the dirty data issue is the one I'm concerned about | 16:03 |
| superdan | sdague: yeah, | 16:04 |
| sdague | and what the internal recovery path is should it trigger | 16:04 |
| sdague | which isn't clear to me | 16:04 |
| ndipanov | lascii, well the idea was to eventually remove device name altogether but that's not gonna happen any time soon | 16:04 |
| superdan | sdague: now that we have gotten to the point of being able to pre-apply the schema, we need to be careful about things that will lock the tables for long periods of time, making it not reasonable to do those at runtime | 16:04 |
| sdague | superdan: ok | 16:05 |
| mriedem | sdague: we raced to -1 it | 16:05 |
| lascii | ndipanov: I'm open to adding a constraint on something else, we would just need to update the queries to use the new tuple | 16:05 |
| sdague | mriedem: :) | 16:05 |
| lascii | ndipanov: I'm not sure what the proper unique way to address a bdm would be | 16:05 |
| mriedem | i found this change last night while going through the asia-pacific hackathon etherpad | 16:05 |
| mriedem | https://etherpad.openstack.org/p/hackathon2_nova_list | 16:06 |
| ndipanov | lascii, but how come we are firing off 2 requests in the first place | 16:06 |
| ndipanov | that's what seems to be the bug to me | 16:06 |
| mriedem | i got to -2 a few things https://etherpad.openstack.org/p/hackathon2_nova_list which was nice :) | 16:06 |
| mriedem | markus_z: lxsli: bauwser: we have an empty trivial bugs list :( https://etherpad.openstack.org/p/liberty-nova-priorities-tracking | 16:07 |
| lascii | ndipanov: every bdm.save() in a cell fires off the request. it seems to happen a few times during the create/attach process | 16:07 |
| bauwser | mriedem: you wanna eat food ? | 16:07 |
| bauwser | mriedem: that's doable | 16:07 |
| mriedem | i will be eating food in about 23 minutes here | 16:07 |
| mriedem | or ordering anyway | 16:07 |
| markus_z | mriedem: thanks for the hint | 16:07 |
| bauwser | lemme grab some pizza slice | 16:07 |
| ndipanov | lascii, I see | 16:08 |
| bauwser | mriedem: but I also have a couple of priority BPs I wanna share | 16:08 |
| bauwser | mriedem: :p | 16:08 |
| phschwartz | mriedem: it is booting now, I will let you know when tests are done | 16:08 |
| bauwser | mriedem: those are a little bit more fat | 16:08 |
| ndipanov | so yes there is one in the api and at least one in the compute service | 16:08 |
| mriedem | bauwser: we're kind of done with bp's now right? | 16:08 |
| bauwser | mriedem: nope | 16:08 |
| bauwser | mriedem: see the etherpad of doom | 16:08 |
| mriedem | i guess you have a couple of days | 16:09 |
| bauwser | mriedem: indeed | 16:09 |
| bauwser | mriedem: guess my mind | 16:09 |
| johnthetubaguy | ndipanov: I wonder if we want bdm attached to instance extra, in some ways? | 16:10 |
| ndipanov | johnthetubaguy, so I sent an email to the thread yesterday that kinda touches on that | 16:10 |
| ndipanov | try to dig it up | 16:10 |
| *** tdurakov has quit IRC | 16:10 | |
| *** rotbart has quit IRC | 16:10 | |
| johnthetubaguy | ndipanov: ah, not caught up on that thread | 16:10 |
| superdan | do we not query by device there? | 16:10 |
| superdan | I thought there were a couple places where we query by volume id, or device name, etc | 16:11 |
| *** leakypipes has quit IRC | 16:11 | |
| *** kiran-r has joined #openstack-nova | 16:13 | |
| johnthetubaguy | superdan: thats true, although I figured we usually already pass in the instance object into those methods, but you are right, needs more digging | 16:13 |
| superdan | maybe, but if we're coming from the volume id, then maybe not | 16:13 |
| *** vilobhmm1 has joined #openstack-nova | 16:14 | |
| ndipanov | so alaski looking at the code | 16:16 |
| ndipanov | that race can happen only if we actually call create() on the object twice | 16:16 |
| ndipanov | and that is most certainly a bug if it is happening | 16:16 |
| ndipanov | save (if it races) should not cause a record to be created afaict | 16:16 |
| lascii | save does call the same cells bdm_update_or_create_at_top method rpcapi method | 16:17 |
| *** dannywilson has joined #openstack-nova | 16:17 | |
| ndipanov | lascii, oops | 16:18 |
| lascii | but it should pass create=False, which it isn't doing | 16:18 |
| lascii | that should also fix this I think | 16:18 |
| ndipanov | I think that's the bug | 16:18 |
| *** alex_klimov has quit IRC | 16:18 | |
| *** claudiub has quit IRC | 16:18 | |
| *** unicell has quit IRC | 16:18 | |
| ndipanov | it passes create = None | 16:19 |
| ndipanov | which causes the update_or_create to be executed | 16:19 |
| lascii | yeah | 16:19 |
| bauwser | question to the audience, shall we treat the small bugfixes having DB upgrades as trivial ? my opinion is no - and IMHO we should also consider them freezed post-FF | 16:19 |
| ndipanov | because that's the default kwarg | 16:19 |
| ndipanov | ooops | 16:19 |
| bauwser | uh, s/freezed/frozen | 16:20 |
| *** 18VAAKDGR is now known as lpetrut1 | 16:20 | |
| superdan | bauwser: no, nothing with a db migration is trivial, IMHO | 16:20 |
| lascii | ndipanov: I still think that an update_or_create method should operate on a uniqe constraint, but that can be deferred for now | 16:20 |
| sdague | bauwser: what superdan said | 16:20 |
| bauwser | superdan: yeah that's my thought | 16:20 |
| sdague | db changes are never counted as trivial | 16:20 |
| ndipanov | lascii, so I'd prefer mdbooth's approach there | 16:20 |
| sdague | because the cost of getting them wrong is super high | 16:20 |
| *** Marga_ has quit IRC | 16:20 | |
| bauwser | superdan: the correlar being that bugfixes having DB migrations can't be considered during RC1, right? | 16:20 |
| bauwser | sdague: yeah that | 16:20 |
| bauwser | sdague: a revert is just super hard | 16:20 |
| lascii | ndipanov: which approach? | 16:21 |
| ndipanov | but maybe I'm wrong | 16:21 |
| sdague | bauwser: well, a revert is actually typically impossible. :) | 16:21 |
| superdan | bauwser: unless it's a CVE or something, yeah | 16:21 |
| bauwser | btw. 'during RC1' means between FF and RC1 tag to me - given our chat last night :) | 16:21 |
| sdague | bauwser: in general yes, unless it's a critical bug | 16:21 |
| bauwser | okay, good to know | 16:21 |
| *** yassine__ has quit IRC | 16:21 | |
| ndipanov | https://review.openstack.org/#/c/141115/13/nova/db/sqlalchemy/api.py | 16:22 |
| ndipanov | lascii, ^ would that work there? | 16:22 |
| openstackgerrit | Dan Smith proposed openstack/nova: Update objects test infrastructure for multiple versions https://review.openstack.org/217818 | 16:24 |
| openstackgerrit | Dan Smith proposed openstack/nova: Add Instance v2.0 object https://review.openstack.org/217819 | 16:24 |
| openstackgerrit | Dan Smith proposed openstack/nova: Make the conductor fixture use version manifests https://review.openstack.org/217851 | 16:24 |
| lascii | ndipanov: I'm not sure. that's for an update but we would need an "insert ... where ..." and I'm not sure how the where should be structured for that | 16:24 |
| *** kiran-r has quit IRC | 16:25 | |
| ndipanov | yeah then a unique constraint... | 16:25 |
| ndipanov | I am not against it really | 16:25 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Fix exception message mistake in WSGI service https://review.openstack.org/215509 | 16:26 |
| superdan | lascii: it'd have to be a subquery I think, which is not awesome | 16:26 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Fix exception message mistake in WSGI service https://review.openstack.org/215509 | 16:26 |
| *** fawadkhaliq has joined #openstack-nova | 16:26 | |
| lascii | superdan: yeah that wouldn't be | 16:27 |
| superdan | adding a constraint on top of an old DB isn't awesome either, of course | 16:27 |
| superdan | but.. | 16:27 |
| ftersin | ndipanov, lascii: how will you decide which thread should win in unique constraint case? | 16:27 |
| superdan | ftersin: the one that wins .. wins | 16:27 |
| ndipanov | ftersin, well it should never happen | 16:28 |
| ndipanov | but we would have caught it sooner if it were blowing up | 16:28 |
| *** hammondr has left #openstack-nova | 16:28 | |
| ndipanov | and not silently creating 2 records | 16:28 |
| superdan | ndipanov: with the UC this race will still happen and one will fail with a constraint error, right? | 16:28 |
| ndipanov | superdan, ^ | 16:28 |
| superdan | so the one that wins the race wins and the other dies | 16:28 |
| *** pkholkin has joined #openstack-nova | 16:28 | |
| lascii | yes | 16:28 |
| *** PaulMurray is now known as ptm_away | 16:28 | |
| ndipanov | well rather raises the exception and we know it's wrong as opposed to just adds a row | 16:29 |
| lascii | although I'm going to add a fix where we shouldn't hit the race | 16:29 |
| superdan | right.. the loser could re-query and apply its updates to the one that was created, | 16:29 |
| markus_z | mriedem: bauwser: I added a few high prio bug patches. I'll look for more on Monday. | 16:29 |
| superdan | but probably better to fail whatever we were doing | 16:29 |
| bauwser | markus_z: will look at them | 16:29 |
| lascii | it would be better to explicitly call create and then explicitly update, rather than multiple calls that can update_or_create | 16:30 |
| markus_z | bauwser: FWIW, http://status.openstack.org/reviews/#nova can be sorted by type since a few days. | 16:30 |
| ftersin | 1st thread puts one record, 2nd one puts changed, if 2nd thread fails and dies, new data will be lost | 16:30 |
| *** tdurakov has joined #openstack-nova | 16:30 | |
| *** Nikolay_St has joined #openstack-nova | 16:30 | |
| mriedem | zzzeek: thanks for looking at https://review.openstack.org/#/c/213584/ | 16:31 |
| ndipanov | lascii, so yes we should have a unique constraint there after some thought and one should be an explicit update | 16:31 |
| superdan | ftersin: that's why I'm saying the create loser can retry as an update, if that makes sense for what is being saved | 16:31 |
| *** annegentle has quit IRC | 16:31 | |
| ndipanov | but also what could happen then is that for whatever reason the update makes it first and hopefully fails | 16:31 |
| zzzeek | mriedem: np :) | 16:31 |
| ndipanov | lascii, so there is still a race there - we really should not return from the create call until we are happy that the BDMs are created | 16:32 |
| ftersin | for retry case, if 1st thread fails, it'll rewrite old data | 16:32 |
| lascii | ndipanov: we could fix that too. just need to convert a cast to a call | 16:32 |
| phschwartz | mriedem: it is running, I have to relocate will get you more info once back online | 16:32 |
| *** browne has quit IRC | 16:32 | |
| bauwser | markus_z: I honestly miss a Gerrit dashboard with that webpage | 16:33 |
| lascii | ndipanov: or add a new rpc call to use | 16:33 |
| ndipanov | lascii, yes | 16:33 |
| *** vilobhmm1 has quit IRC | 16:33 | |
| ndipanov | so then create in the cpu cell api needs to block until it knows it's done | 16:33 |
| bauwser | markus_z: because it doesn't tell me if the bug is actionable or not | 16:33 |
| *** claudiub has joined #openstack-nova | 16:33 | |
| ndipanov | otherwise with a UC we will be consistent but still risk failing due to weird races | 16:34 |
| *** tonytan4ever has quit IRC | 16:34 | |
| *** exploreshaifali has joined #openstack-nova | 16:34 | |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: objects: Hook migration object into Instance https://review.openstack.org/216733 | 16:35 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: Claims: Make sure move claims crate a migration context records https://review.openstack.org/218385 | 16:35 |
| *** jistr has quit IRC | 16:35 | |
| ftersin | do you understand where from the first call comes, and why is it async? | 16:35 |
| lascii | ndipanov: agreed | 16:36 |
| *** derekh has quit IRC | 16:36 | |
| *** salv-orlando has quit IRC | 16:36 | |
| markus_z | bauwser: When do you consider it as actionable? | 16:36 |
| bauwser | markus_z: checking the bug state is not enough | 16:37 |
| lascii | ftersin: I know where the first call comes from. I don't know why it was done as an async call | 16:37 |
| bauwser | markus_z: because it can be in progress but with abandoned changes | 16:37 |
| lascii | ftersin: but I have a few fixes that I'll work on after I get some lunch | 16:37 |
| bauwser | markus_z: also, some bugfix can be -W'd or -2'd | 16:37 |
| ftersin | lascii: from api service? | 16:37 |
| *** mriedem is now known as mriedem_lunch | 16:38 | |
| lascii | ftersin: they're all coming from the compute. the initial create in the api is blocked for cells | 16:38 |
| *** yamahata has quit IRC | 16:38 | |
| bauwser | lascii: I obviously know you're pretty busy with that bug, but could you take time to sneak peek at the ReqSpec object so we could try to land a bit more changes before FF ? | 16:39 |
| markus_z | bauwser: In the last column on the right hand side is the summary of the patch set. I thought this reflects what you asked for, doesn't it? | 16:39 |
| lascii | bauwser: sure, I'll take a look after | 16:39 |
| bauwser | lascii: sorry about that, I hate pinging that way | 16:39 |
| bauwser | lascii: it's just all about deadlines y'know :( | 16:40 |
| lascii | bauwser: no worries | 16:40 |
| *** annegentle has joined #openstack-nova | 16:40 | |
| bauwser | markus_z: indeed | 16:40 |
| bauwser | markus_z: but I can't just refine | 16:40 |
| *** earlephilhower has joined #openstack-nova | 16:40 | |
| *** penick has joined #openstack-nova | 16:41 | |
| bauwser | markus_z: and have a fine-grained query like label:Verified<=-1,jenkins AND label:Code-Review>=0 | 16:41 |
| *** dprince has quit IRC | 16:41 | |
| ftersin | lascii: i didn't find a code on compute which can make the first call, could you please point me on? | 16:42 |
| bauwser | ftersin: it goes to the cells api | 16:42 |
| ndipanov | so superdan tell me if I am mistaken since you messed with that code recently | 16:42 |
| ndipanov | revert_resize and finish_revert_resize | 16:42 |
| superdan | ndipanov: um, I dunno, maybe? | 16:42 |
| ndipanov | so you swap back to the old flavor in finish_revert_resize (on the orriginal host) | 16:43 |
| ndipanov | but we release the claim in revert_resize - | 16:43 |
| ndipanov | whihc is wrong because at that point we start accounting the new flavor on the old node | 16:44 |
| ndipanov | does that make sense? | 16:44 |
| *** jwcroppe has joined #openstack-nova | 16:44 | |
| *** unicell has joined #openstack-nova | 16:46 | |
| superdan | ndipanov: ...this is my blank stare face... | 16:46 |
| ndipanov | lol ok nevermind | 16:47 |
| superdan | I don't think _I_ have changed any of that, at least not recently | 16:47 |
| ndipanov | superdan, well no you did change how flavors are swapped | 16:47 |
| ndipanov | so I thought maybe you looked under the hood a bit | 16:47 |
| superdan | because I was doing the flavor conversion, or I actually changed the logic? | 16:47 |
| ndipanov | flavor conversion | 16:48 |
| ndipanov | but you actually only changed the method | 16:49 |
| ndipanov | call sites were not touched so in that case... | 16:49 |
| ndipanov | nevermind :) | 16:49 |
| superdan | heh, okay | 16:49 |
| *** jogo is now known as flashgordon | 16:49 | |
| bauwser | so, last call for a very nice feature having +2 on all the line https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/allocation-ratio-to-resource-tracker,n,z | 16:51 |
| superdan | bauwser: I'll hit those in a bit | 16:51 |
| bauwser | (gosh, I hate deadlines) | 16:51 |
| *** Apoorva has joined #openstack-nova | 16:51 | |
| bauwser | superdan: no worries | 16:51 |
| ndipanov | ugh so I think I found another bug | 16:51 |
| bauwser | superdan: I'm not super excited to hassle people | 16:51 |
| ndipanov | I need to test this tomorrow | 16:51 |
| ndipanov | or whenever | 16:52 |
| superdan | bauwser: don't lie, I know you love it | 16:52 |
| *** tjones has left #openstack-nova | 16:52 | |
| *** Apoorva has quit IRC | 16:52 | |
| bauwser | superdan: eh, I'm French, so that's probably true | 16:52 |
| *** Apoorva has joined #openstack-nova | 16:53 | |
| superdan | heh | 16:55 |
| *** Marga_ has joined #openstack-nova | 16:56 | |
| superdan | ndipanov: is your -1 still valid here? https://review.openstack.org/#/c/216362/5 | 16:59 |
| superdan | I think not, but just want to make sure | 16:59 |
| *** dprince has joined #openstack-nova | 16:59 | |
| *** mtanino has quit IRC | 17:00 | |
| ndipanov | yeah not but haven't had time to review | 17:00 |
| *** haomaiwang has quit IRC | 17:01 | |
| ndipanov | will just remove it | 17:01 |
| superdan | I can just +W, was just checking | 17:01 |
| superdan | the config defaulting thing has been addressed | 17:01 |
| ndipanov | nice | 17:01 |
| *** lucasagomes is now known as lucas-beer | 17:01 | |
| *** haomaiwang has joined #openstack-nova | 17:02 | |
| superdan | bauwser: I think this is -2: https://review.openstack.org/#/c/216696/5 | 17:03 |
| superdan | bauwser: we can't do that until M, right? | 17:03 |
| superdan | people that deploy releases will not be able to roll compute nodes to the commit before this one before they apply this one to the controllers | 17:04 |
| superdan | johnthetubaguy: ^ | 17:05 |
| bauwser | superdan: mmm | 17:06 |
| ndipanov | why didn't we just introduce new config options again | 17:06 |
| ndipanov | ? | 17:06 |
| superdan | maybe not | 17:07 |
| *** nic has joined #openstack-nova | 17:07 | |
| ndipanov | for computes | 17:07 |
| superdan | now I'm just not sure why you did it in this order | 17:07 |
| ndipanov | and deprecated the one in the shceuler | 17:07 |
| ndipanov | this is just confusing really | 17:07 |
| bauwser | superdan: it should not change | 17:07 |
| ndipanov | we are changing what that option means | 17:07 |
| superdan | if we did the computenode thing first, then we wouldn't have to do the temporary hack on the scheduler filters, right? | 17:07 |
| ndipanov | it should be a new one | 17:07 |
| ndipanov | </rant> | 17:07 |
| superdan | ndipanov: that's one way to look at it, but I think you could be less strict about it | 17:08 |
| bauwser | superdan: the filters are now using the RT options | 17:08 |
| bauwser | superdan: so since we modify those RT opt values, I provided the first change | 17:08 |
| *** mjturek1 has quit IRC | 17:08 | |
| superdan | bauwser: right, but if you did things in this order: | 17:09 |
| bauwser | superdan: when updating the scheduler, it will use the host_state metrics but going to the ComputeNode object facade,it will read the sched opts actually | 17:09 |
| superdan | 1. Add the support in the DB, object, defaulting to the values when 0.0 | 17:09 |
| superdan | 2. Make the scheduler filters use the value from the object (which will be defaulted properly) | 17:09 |
| superdan | 3. Make the compute nodes report their values | 17:09 |
| superdan | then you wouldn't need this hack, right? | 17:09 |
| bauwser | superdan: the hack from the filters ? | 17:10 |
| superdan | yeah | 17:10 |
| bauwser | superdan: I see | 17:10 |
| bauwser | superdan: yeah, you're right | 17:10 |
| *** erhudy has joined #openstack-nova | 17:10 | |
| bauwser | superdan: okay, I can send a new version of all of htat | 17:10 |
| superdan | well, | 17:10 |
| bauwser | that was basically johnthetubaguy's thoughts | 17:11 |
| superdan | this is already in the gate, so here's what I think we should do: | 17:11 |
| superdan | I'll +W this and if it all lands, then that's fine | 17:11 |
| superdan | if it doesn't, then I think it would be cleaner to just drop the first patch and reorder | 17:11 |
| superdan | is that okay? | 17:11 |
| bauwser | superdan: agreed | 17:12 |
| superdan | it's 123rd in the gate, so I think it's nearly impossible that it will land, so get ready :) | 17:12 |
| bauwser | superdan: ack | 17:12 |
| *** salv-orlando has joined #openstack-nova | 17:12 | |
| superdan | I don't think this is enough of a rush to need to work on this over the weekend, but if you do, ping me and I'll hit the updated version then | 17:12 |
| *** vilobhmm1 has joined #openstack-nova | 17:13 | |
| bauwser | superdan: it should be okay since the gate is quite empty by the EUmornings | 17:13 |
| johnthetubaguy | bauwser: it was mental this morning | 17:13 |
| johnthetubaguy | hoping its not on monday morning | 17:13 |
| bauwser | johnthetubaguy: oh the gate | 17:13 |
| *** markus_z has quit IRC | 17:13 | |
| johnthetubaguy | yup | 17:13 |
| bauwser | right, I was only rechecking this morning | 17:13 |
| bauwser | superdan: if you prefer, I can abandon those changes and restore them, that should stop gating | 17:14 |
| *** yamahata has joined #openstack-nova | 17:14 | |
| bauwser | superdan: and then send a new version if that feels much better | 17:15 |
| superdan | bauwser: stopping the gate isn't hard, just changing the commit message on the first one will do it | 17:15 |
| lascii | ftersin: the bdm is created in the cell at https://github.com/openstack/nova/blob/master/nova/cells/scheduler.py#L114 | 17:15 |
| johnthetubaguy | superdan: I am wondering if we want to do the "only merge blueprint patches" thing from Monday, till we get tagged on Thursday? | 17:15 |
| superdan | bauwser: johnthetubaguy: do you feel strongly about doing it the cleaner way? if so, lets just dump it out of the gate sooner than later | 17:15 |
| bauwser | superdan: okay, lemme do that, I acutally prefer the cleaner way | 17:15 |
| superdan | bauwser: okay | 17:15 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Fix the default allocation ratios to 0.0 - ABANDON https://review.openstack.org/217737 | 17:16 |
| johnthetubaguy | I didn't block on that, because I was feeling bad about blocking on the other stuff, but yeah, I would love the cleaner way | 17:16 |
| bauwser | should be done in the next 10 mins | 17:16 |
| superdan | okay | 17:17 |
| superdan | bauwser: sorry to derail further | 17:17 |
| superdan | but thanks for preferring cleanliness :) | 17:17 |
| bauwser | superdan: nah you definitely made a good point | 17:17 |
| *** danpb has quit IRC | 17:17 | |
| ftersin | lascii: ah, other module. thanks! | 17:17 |
| bauwser | superdan: I by far prefer to delay my stuff than putting crap | 17:18 |
| lascii | ftersin: np. technically it happens on the nova-cell service in the cell before it goes to the compute | 17:18 |
| superdan | me too | 17:18 |
| *** lpetrut1 has quit IRC | 17:19 | |
| *** penick has quit IRC | 17:19 | |
| *** aix has quit IRC | 17:20 | |
| phschwartz | mriedem_lunch: The run crashed in the middle with an issue unrelated to your review. rerunning now. | 17:21 |
| *** wendar_ is now known as wendar | 17:23 | |
| *** lucas-beer has quit IRC | 17:25 | |
| *** Marga_ has quit IRC | 17:26 | |
| *** sachin has joined #openstack-nova | 17:29 | |
| *** gszasz has quit IRC | 17:32 | |
| *** changbl has quit IRC | 17:33 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds RDPConsoleOps unit tests https://review.openstack.org/139798 | 17:34 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds HyperVDriver unit tests https://review.openstack.org/148980 | 17:34 |
| *** browne has joined #openstack-nova | 17:37 | |
| *** SourabhP has joined #openstack-nova | 17:38 | |
| *** mriedem_lunch is now known as mriedem | 17:40 | |
| mriedem | phschwartz: ok | 17:40 |
| phschwartz | mriedem: the run is going so hopefully I don't get a kernel panic again this time. lol | 17:41 |
| *** rdopiera has quit IRC | 17:43 | |
| *** Marga_ has joined #openstack-nova | 17:44 | |
| *** jaypipes has joined #openstack-nova | 17:46 | |
| jaypipes | bauwser: on https://review.openstack.org/#/c/213590/, do you still have objections on that? | 17:47 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Add cpu_allocation_ratio and ram_allocation_ratio to ComputeNode https://review.openstack.org/215471 | 17:47 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Update HostManager and filters to use ComputeNode ratios https://review.openstack.org/216696 | 17:47 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Update ComputeNode values with allocation ratios in the RT https://review.openstack.org/216362 | 17:47 |
| bauwser | jaypipes: yeah sorry, had no time to put my -1 again :( | 17:48 |
| bauwser | jaypipes: but my -1 is actually for docstrings | 17:48 |
| bauwser | jaypipes: since I only wanted to rebase as I wasn't discussing with you before that | 17:48 |
| bauwser | superdan: johnthetubaguy: the allocation-ratio changes are now updated | 17:49 |
| bauwser | superdan: johnthetubaguy: https://review.openstack.org/215471 take your time to doublecheck before +2/+W if you wish | 17:49 |
| superdan | bauwser: yep, looking now | 17:49 |
| bauwser | superdan: johnthetubaguy: I basically just reordered the changes and modified where I was updating the RT opts | 17:50 |
| superdan | johnthetubaguy: are you around to check or do you want me to just check them over and fast approve? | 17:50 |
| *** snikitin has joined #openstack-nova | 17:50 | |
| superdan | bauwser: you intended to remove the warning from these? https://review.openstack.org/#/c/216362/6/nova/compute/resource_tracker.py,cm | 17:51 |
| bauwser | superdan: *facepalm | 17:51 |
| bauwser | superdan: no, it's a miss | 17:51 |
| bauwser | superdan: I'll put a NOTE tho as per the comments | 17:52 |
| superdan | bauwser: okay, I think it should be just something like ... yeah, NOTE :) | 17:52 |
| *** jwcroppe has quit IRC | 17:52 | |
| superdan | aside from that, gerrit says they're basically exactly the same | 17:52 |
| superdan | so if johnthetubaguy isn't around by the time you fix that NOTE, I'll just re-+W them | 17:52 |
| bauwser | superdan: happy with the docstring I made in the opt - apart of the WARNING/NOTE ? | 17:53 |
| *** VW_ has quit IRC | 17:53 | |
| bauwser | superdan: I should maybe be more verbose and explain that having 0.0 doesn't mean that you'll get that | 17:53 |
| johnthetubaguy | superdan: sorry, I am still around, just, I think fast approve is fine | 17:53 |
| *** VW_ has joined #openstack-nova | 17:54 | |
| superdan | johnthetubaguy: okay | 17:54 |
| superdan | bauwser: well, it really should be something like this I think: | 17:55 |
| superdan | "NOTE: This can be set per-compute, or if set to 0.0, the value set on the scheduler node(s) will be used" | 17:55 |
| superdan | right? | 17:55 |
| bauwser | superdan: well, it's confusing | 17:56 |
| bauwser | superdan: because that text will be provided also on the scheduler | 17:56 |
| bauwser | superdan: here is my try | 17:56 |
| superdan | sure, but that seems fine | 17:56 |
| *** achanda has joined #openstack-nova | 17:56 | |
| bauwser | NOTE: This can be set per-compute, or if set to 0.0, the value set on the scheduler node(s) will be used and defaulted to 16.0" | 17:58 |
| superdan | ah, okay, sure | 17:58 |
| *** unicell has quit IRC | 17:58 | |
| *** VW_ has quit IRC | 17:58 | |
| *** unicell1 has joined #openstack-nova | 17:58 | |
| superdan | and then the "defaulted to 16" part can be removed in mitaka, when we *actually* set the default to 16, yes? | 17:59 |
| bauwser | I just want operators to make sure that if they haven't set a value on the scheduler, they can keep the 0.0 value because it will default to 16 | 17:59 |
| superdan | yeah | 17:59 |
| superdan | that's cool | 17:59 |
| bauwser | superdan: zactly | 17:59 |
| superdan | okay, push it up already :) | 17:59 |
| *** VW_ has joined #openstack-nova | 17:59 | |
| *** electroc_ has quit IRC | 18:00 | |
| *** haomaiwang has quit IRC | 18:01 | |
| *** 20WABZC7L has joined #openstack-nova | 18:02 | |
| *** Marga_ has quit IRC | 18:03 | |
| bauwser | oh man, a git review -y is just taking a while here | 18:03 |
| * bauwser thanks Netflix and my local DSL provider who is unable to give me more than 7Mbps | 18:04 | |
| *** mjturek1 has joined #openstack-nova | 18:04 | |
| superdan | traffic shaping! | 18:04 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Update ComputeNode values with allocation ratios in the RT https://review.openstack.org/216362 | 18:04 |
| bauwser | hurrah, 4 mins for that... | 18:05 |
| johnthetubaguy | the A bit of ADSL tends to suck for that kind of thing | 18:05 |
| johnthetubaguy | I can give that a quick look | 18:05 |
| bauwser | johnthetubaguy: what sucks is how ISPs just stopped to deploy FC here | 18:05 |
| superdan | jaypipes -1d | 18:05 |
| superdan | I think he's missed some of the previous discussion | 18:05 |
| bauwser | orly ? | 18:05 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Add a warning in the microversion docs around the usage of 'latest' https://review.openstack.org/218403 | 18:06 |
| bauwser | okay, I see his comments, replying | 18:07 |
| *** tjones has joined #openstack-nova | 18:11 | |
| *** mmedvede has quit IRC | 18:11 | |
| *** marzif has joined #openstack-nova | 18:12 | |
| mriedem | ctrath: sounds like a backlog spec is going to be coming for this in mitaka http://lists.openstack.org/pipermail/openstack-dev/2015-August/073119.html | 18:12 |
| mriedem | if you're looking to throw bodies on it | 18:12 |
| superdan | he's on a plane I think | 18:13 |
| superdan | hopefully he'll see the replies | 18:13 |
| mriedem | api change in nova, novaclient cli change, probably openstackclient change, and horizon change for doug-fish | 18:13 |
| ctrath | mriedem: thanks | 18:13 |
| *** aix has joined #openstack-nova | 18:14 | |
| ctrath | rlrooster or ccarmack: looking for something to do? | 18:14 |
| rlrooster | ctrath: mriedem: would this require a bp? (and I assume it's a not-until-mitaka thing) | 18:19 |
| mriedem | yes | 18:19 |
| mriedem | i assume tim bell will post a backlog spec with his requirement | 18:20 |
| openstackgerrit | Jay Pipes proposed openstack/nova: rt: Rewrite abort and update_usage tests https://review.openstack.org/213590 | 18:20 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Cleanup RT _instance_in_resize_state() https://review.openstack.org/213589 | 18:20 |
| *** exploreshaifali has quit IRC | 18:21 | |
| bauwser | superdan: thanks for your help, heading out | 18:21 |
| superdan | bauwser: np, enjoy the weekend | 18:21 |
| jaypipes | bauwser, superdan: would really appreciate re-reviews on patch series starting at https://review.openstack.org/#/c/213589/ | 18:21 |
| bauwser | superdan: I'll be around in the next hours | 18:21 |
| superdan | jaypipes: okay | 18:22 |
| superdan | jaypipes: see our comments on bauwser's patches? | 18:22 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Move test of claim context manager to test_tracker https://review.openstack.org/213591 | 18:22 |
| bauwser | superdan: just deferring a few hours to family | 18:22 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Move resize claim abort test to test_tracker https://review.openstack.org/213592 | 18:22 |
| *** mtanino has joined #openstack-nova | 18:23 | |
| jaypipes | bauwser: imbricated... I don't think that means what you think it means :) | 18:23 |
| figleaf | bauwser: Family??? Where are your priorities, man? :) | 18:23 |
| mriedem | belliott: nice work https://review.openstack.org/#/c/218362/ | 18:25 |
| mriedem | a few things inline | 18:25 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Add a warning in the microversion docs around the usage of 'latest' https://review.openstack.org/218403 | 18:27 |
| *** electrocucaracha has joined #openstack-nova | 18:27 | |
| *** electrocucaracha has quit IRC | 18:28 | |
| *** vladikr has quit IRC | 18:30 | |
| *** amotoki has quit IRC | 18:31 | |
| *** lyanchih has quit IRC | 18:32 | |
| *** fawadkhaliq has quit IRC | 18:33 | |
| *** lyanchih has joined #openstack-nova | 18:33 | |
| *** electrocucaracha has joined #openstack-nova | 18:34 | |
| *** VW_ has quit IRC | 18:36 | |
| *** jaypipes has quit IRC | 18:40 | |
| *** snikitin has quit IRC | 18:41 | |
| *** tdurakov has quit IRC | 18:41 | |
| *** snikitin has joined #openstack-nova | 18:42 | |
| *** tdurakov has joined #openstack-nova | 18:42 | |
| *** tonytan4ever has joined #openstack-nova | 18:42 | |
| *** achanda has quit IRC | 18:44 | |
| *** exploreshaifali has joined #openstack-nova | 18:47 | |
| phschwartz | mriedem: ok, looking good. testing ran through twice. Only errors I got were related to thirdparty tools as I shorted installing boto ;) | 18:49 |
| mriedem | phschwartz: ok, we're -1 on the change for now anyway, but good to know you're not finding issues | 18:50 |
| belliott | mriedem: cool, thx for comments | 18:52 |
| *** xyang1 has joined #openstack-nova | 18:52 | |
| *** promulo has joined #openstack-nova | 18:58 | |
| openstackgerrit | Aditi Rajagopal proposed openstack/nova: Rm openstack/common/versionutils from setup.cfg https://review.openstack.org/217924 | 18:59 |
| *** jcoufal has quit IRC | 18:59 | |
| *** claudiub has quit IRC | 18:59 | |
| *** vivekd_ has joined #openstack-nova | 18:59 | |
| *** tjones has left #openstack-nova | 19:00 | |
| *** 20WABZC7L has quit IRC | 19:01 | |
| *** vivekd has quit IRC | 19:01 | |
| *** vivekd_ is now known as vivekd | 19:01 | |
| *** haomaiwang has joined #openstack-nova | 19:02 | |
| *** dannywilson has quit IRC | 19:09 | |
| *** achanda has joined #openstack-nova | 19:11 | |
| *** nic has quit IRC | 19:14 | |
| *** nic has joined #openstack-nova | 19:15 | |
| *** dave-mccowan has quit IRC | 19:16 | |
| mriedem | ccarmack: need to fix up the pep8 failure here https://review.openstack.org/#/c/217840/ | 19:16 |
| *** lyanchih has quit IRC | 19:17 | |
| mriedem | then i'm +2 | 19:17 |
| *** promulo has quit IRC | 19:17 | |
| *** dannywilson has joined #openstack-nova | 19:17 | |
| *** promulo has joined #openstack-nova | 19:18 | |
| *** promulo has quit IRC | 19:18 | |
| mriedem | mrda: are you around? | 19:19 |
| figleaf | /away errands | 19:20 |
| figleaf | doh! | 19:20 |
| mriedem | or anyone else that cares about 1:M support for vcenter or ironic, | 19:20 |
| mriedem | https://review.openstack.org/#/c/211083/ | 19:20 |
| mriedem | superdan: ^ seems that's something we wouldn't want to enable/support, | 19:20 |
| mriedem | given the patch landed to not allow 1:M for vcenter, | 19:20 |
| mriedem | he has a point about ironic, but i'd think we'd want to restrict that at some point too | 19:21 |
| superdan | it's not right at all | 19:22 |
| superdan | you can't add up all the disk for all your nodes | 19:22 |
| superdan | because you can't actually schedule a 4T instance | 19:22 |
| mriedem | but i want to | 19:22 |
| mriedem | and i have a mainframe | 19:22 |
| superdan | let me rephrase | 19:23 |
| mriedem | heh | 19:23 |
| superdan | because you can't actually schedule a 4T instance because you have 4 1T nodes | 19:23 |
| * mriedem will enjoy this | 19:23 | |
| mriedem | so i changed my -2 to a -1, but it smells like a -2 again | 19:23 |
| mriedem | feel free to drop that hammer if you lke | 19:23 |
| mriedem | *like | 19:24 |
| superdan | I'll do the -2 | 19:24 |
| mtreinish | mriedem: heh, that was a great response | 19:24 |
| mtreinish | mriedem: now I'm going to bug you for an account on your mainframe | 19:24 |
| mriedem | excellent | 19:24 |
| *** Apoorva_ has joined #openstack-nova | 19:25 | |
| mtreinish | because I couldn't even get access to one when I was working on linux nfs interop with z/os | 19:25 |
| mriedem | i plan on running https://upload.wikimedia.org/wikipedia/en/2/2b/Eye_of_the_Beholder_I_PC_box.jpeg on it | 19:25 |
| *** Apoorva has quit IRC | 19:25 | |
| *** yamahata has quit IRC | 19:26 | |
| superdan | mtreinish: I actually had one in machine-level debug for a while | 19:27 |
| superdan | and I crashed z/VM and they had to actually reboot | 19:27 |
| superdan | which apparently was something they had to look up how to do | 19:27 |
| mriedem | superdan: running divide-by-zero on the mainframe | 19:29 |
| mriedem | so, oslo lib integration would not be considered something we let in post non-priority feature freeze right? | 19:29 |
| superdan | heh, no I think they protect against that :) | 19:29 |
| mriedem | like at this point, i'm thinking integration of new oslo libs is not kosher | 19:30 |
| superdan | agreed | 19:30 |
| mtreinish | superdan: heh, I just have this image in my head of a guy banging 2 rocks together to get a spark as the reboot procedure | 19:33 |
| dimsum__ | ++ it can wait | 19:33 |
| dimsum__ | mriedem: ++ it can wait | 19:33 |
| superdan | mriedem: may have been, I didn't get to watch :) | 19:34 |
| mriedem | dimsum__: cool | 19:34 |
| dimsum__ | mriedem: oslo.cache i presume | 19:34 |
| mriedem | yeah | 19:34 |
| mriedem | not comfortable with landing that at this point | 19:34 |
| dimsum__ | mriedem: i am with you | 19:34 |
| mriedem | dimsum__: hold me | 19:34 |
| dimsum__ | :) | 19:34 |
| mriedem | oh sorry, in channel... | 19:34 |
| dimsum__ | lol | 19:34 |
| rlrooster | this is getting weird | 19:35 |
| *** lbeliveau has quit IRC | 19:35 | |
| mriedem | if this is weird, your bar is in the wrong place | 19:36 |
| superdan | getting weird means .. not there yet | 19:36 |
| superdan | so his bar may be positioned correctly | 19:36 |
| superdan | but he's just warning us ahead of time | 19:36 |
| *** changbl has joined #openstack-nova | 19:37 | |
| rlrooster | the only bar I care about is the one downtown with a nice cold brewski | 19:37 |
| mriedem | hi-o | 19:37 |
| *** devlaps has joined #openstack-nova | 19:37 | |
| *** IanGovett has quit IRC | 19:39 | |
| *** Apoorva_ has quit IRC | 19:41 | |
| *** rushil has joined #openstack-nova | 19:42 | |
| *** Piet has joined #openstack-nova | 19:42 | |
| *** achanda has quit IRC | 19:43 | |
| *** eharney has quit IRC | 19:43 | |
| *** thangp has quit IRC | 19:46 | |
| *** Apoorva has joined #openstack-nova | 19:47 | |
| *** rfolco has quit IRC | 19:47 | |
| *** achanda has joined #openstack-nova | 19:47 | |
| *** dave-mccowan has joined #openstack-nova | 19:47 | |
| openstackgerrit | Chuck Carmack proposed openstack/nova: Don't trace on InstanceInfoCacheNotFound when refreshing network info_cache https://review.openstack.org/217840 | 19:49 |
| ccarmack | mriedem: ^, next time I'll run pep8 first | 19:50 |
| *** Apoorva has quit IRC | 19:51 | |
| *** jcoufal has joined #openstack-nova | 19:52 | |
| *** cfarquhar has quit IRC | 19:52 | |
| openstackgerrit | j_king proposed openstack/nova: libvirt: Add wait for vif unplug events on destroy https://review.openstack.org/210126 | 19:53 |
| *** rushil has quit IRC | 19:54 | |
| *** rushil has joined #openstack-nova | 19:54 | |
| *** cfarquhar has joined #openstack-nova | 19:55 | |
| *** cfarquhar has quit IRC | 19:55 | |
| *** cfarquhar has joined #openstack-nova | 19:55 | |
| *** jichen has joined #openstack-nova | 19:55 | |
| *** vivekd_ has joined #openstack-nova | 19:55 | |
| *** lpetrut has joined #openstack-nova | 19:56 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: Cells: Improve block device mapping update/create calls https://review.openstack.org/218450 | 19:58 |
| *** tjones has joined #openstack-nova | 19:59 | |
| *** vivekd has quit IRC | 19:59 | |
| *** vivekd_ is now known as vivekd | 19:59 | |
| mriedem | mtreinish: ken'ichi was feeling kind https://review.openstack.org/#/c/210684/ | 20:00 |
| *** haomaiwang has quit IRC | 20:01 | |
| mtreinish | heh | 20:01 |
| openstackgerrit | Brian Elliott proposed openstack/nova: Add migration policy to upgrades devref https://review.openstack.org/218362 | 20:01 |
| * bauwser waves back | 20:01 | |
| mtreinish | mriedem: I think he was just calling you strict | 20:01 |
| *** haomaiwang has joined #openstack-nova | 20:02 | |
| dimsum__ | bauwser: get some sleep :) | 20:04 |
| *** penick has joined #openstack-nova | 20:04 | |
| bauwser | dimsum__: what? | 20:05 |
| bauwser | dimsum__: sleep? | 20:05 |
| dimsum__ | bauwser: isn't it late for you? | 20:05 |
| *** penick_ has joined #openstack-nova | 20:05 | |
| bauwser | dimsum__: meh, not so late that for the Late Showsd | 20:06 |
| bauwser | Late Shows | 20:06 |
| bauwser | not so late than for (even) | 20:06 |
| dimsum__ | :) | 20:06 |
| *** eglynn has quit IRC | 20:08 | |
| *** penick has quit IRC | 20:09 | |
| *** penick_ is now known as penick | 20:09 | |
| mriedem | ha | 20:09 |
| mriedem | i think the point has been proven | 20:09 |
| *** rlrooster has left #openstack-nova | 20:12 | |
| *** rushil has quit IRC | 20:13 | |
| *** Apoorva has joined #openstack-nova | 20:14 | |
| vishwanathj | Hello, I have a question related to cirros cloud-init enabled images? Is this the right forum to ask those questions? Thanks | 20:18 |
| mriedem | vishwanathj: see channel topic | 20:19 |
| *** sachin has quit IRC | 20:19 | |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: Claims: Make sure move claims crate a migration context records https://review.openstack.org/218385 | 20:20 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: objects: Hook migration object into Instance https://review.openstack.org/216733 | 20:20 |
| *** sachin has joined #openstack-nova | 20:20 | |
| vishwanathj | mriedem, thanks, looks like i need to use the channel openstack for nova deployment questions | 20:22 |
| *** smatzek has quit IRC | 20:26 | |
| *** exploreshaifali has quit IRC | 20:28 | |
| *** achanda has quit IRC | 20:28 | |
| *** vivekd has quit IRC | 20:29 | |
| *** Nikolay_St has quit IRC | 20:29 | |
| *** ccarmack has left #openstack-nova | 20:32 | |
| *** matrohon has joined #openstack-nova | 20:34 | |
| *** apuimedo_ has joined #openstack-nova | 20:35 | |
| mriedem | lascii: you might be interested in this at some point, i'm trying to think of any alternative approach https://review.openstack.org/#/c/187145/ | 20:35 |
| *** lpetrut has quit IRC | 20:36 | |
| *** tjones has quit IRC | 20:36 | |
| *** tjones has joined #openstack-nova | 20:38 | |
| *** eglynn has joined #openstack-nova | 20:40 | |
| lascii | mriedem: it does seem like that should raise an exception for the decorator to handle | 20:41 |
| lascii | I think it's returning a status due to the hook though | 20:41 |
| mriedem | ok, i was wondering where build_results was read | 20:42 |
| mriedem | didn't think about the hook | 20:42 |
| *** sachin has quit IRC | 20:42 | |
| *** sachin has joined #openstack-nova | 20:43 | |
| *** dustins_ has joined #openstack-nova | 20:44 | |
| *** dustins has quit IRC | 20:45 | |
| *** dustins_ is now known as dustins | 20:45 | |
| lascii | we could make that rabbit hole deeper by having another inner method with the decorator, and then catch the exceptions and convert them to a status for the hook. that's ugly as well though | 20:46 |
| mriedem | ew | 20:46 |
| mriedem | yeah | 20:46 |
| mriedem | i'm not crazy about this change but i think it's probably ok | 20:47 |
| mriedem | it's definitely better than not being able to delete an instance that failed to build | 20:47 |
| *** sachin has quit IRC | 20:47 | |
| lascii | the task state prevents a delete? | 20:47 |
| mriedem | yes | 20:47 |
| mriedem | that's what the reset-state API is for, but it's admin-only by dfeault | 20:47 |
| mriedem | *default | 20:47 |
| lascii | huh, I thought delete was always supposed to work | 20:48 |
| mriedem | http://git.openstack.org/cgit/openstack/nova/tree/nova/compute/api.py#n1900 | 20:48 |
| mriedem | hells naw | 20:48 |
| lascii | but yeah, the change seems reasonable to me | 20:48 |
| *** sachin has joined #openstack-nova | 20:48 | |
| mriedem | i guess force-delete doesn't check vm_state or task_state http://git.openstack.org/cgit/openstack/nova/tree/nova/compute/api.py#n1938 | 20:49 |
| *** boden has joined #openstack-nova | 20:49 | |
| lascii | looks like that's in an extension most people may not have loaded | 20:49 |
| *** krtaylor has quit IRC | 20:51 | |
| lascii | oh, vm_state=None, task_state=None means allow all | 20:51 |
| *** sachin has quit IRC | 20:52 | |
| mriedem | yeah i guess i thought i remember there being a problem where force-delete doesn't work | 20:52 |
| mriedem | maybe superdan remembers | 20:52 |
| mriedem | this was related https://review.openstack.org/#/c/163798/ | 20:53 |
| superdan | yeah, I think vm_state=[None] means it has to be None, | 20:53 |
| superdan | =None means anything | 20:53 |
| *** sachin has joined #openstack-nova | 20:53 | |
| ndipanov | mriedem, superdan https://review.openstack.org/218201 slam dunk | 20:54 |
| mriedem | hells bells | 20:54 |
| mriedem | we should just create an ANY sentinel | 20:54 |
| mriedem | because None and [None] is confusing | 20:54 |
| lascii | yeah, and the default is vm_state=None, task_state=(None,) | 20:54 |
| mriedem | ndipanov: i still remember when the bush administration said iraq was a slam dunk | 20:54 |
| *** sachin has quit IRC | 20:55 | |
| lascii | didn't you see the banner though, mission accomplished | 20:55 |
| superdan | ndipanov: johnthetubaguy has a comment on there | 20:55 |
| *** jaypipes has joined #openstack-nova | 20:55 | |
| mriedem | lascii: i remember | 20:55 |
| ndipanov | yeah just responding missed that | 20:55 |
| mriedem | *huge balls* | 20:55 |
| mriedem | texas-sized balls | 20:55 |
| jaypipes | superdan: hey, I'm back... :) | 20:55 |
| *** sachin has joined #openstack-nova | 20:56 | |
| *** sachin has quit IRC | 20:57 | |
| *** matrohon has quit IRC | 20:58 | |
| superdan | jaypipes: in EDT? | 20:58 |
| *** sachin has joined #openstack-nova | 20:59 | |
| *** dprince has quit IRC | 21:00 | |
| *** sachin has quit IRC | 21:00 | |
| *** haomaiwang has quit IRC | 21:01 | |
| *** tjones has left #openstack-nova | 21:01 | |
| *** sachin has joined #openstack-nova | 21:01 | |
| openstackgerrit | Chuck Carmack proposed openstack/nova: Make the server group invalid format message more verbose https://review.openstack.org/218468 | 21:01 |
| *** sachin has quit IRC | 21:01 | |
| *** haomaiwang has joined #openstack-nova | 21:01 | |
| *** SourabhP has quit IRC | 21:01 | |
| *** SourabhP has joined #openstack-nova | 21:01 | |
| *** sachin has joined #openstack-nova | 21:01 | |
| *** raildo is now known as raildo-afk | 21:02 | |
| bauwser | lascii: I'm just looking at implementing a catch exception for http://logs.openstack.org/17/213317/15/check/gate-tempest-dsvm-cells/9693a2c/logs/screen-n-cell-region.txt.gz?level=TRACE#_2015-08-28_19_27_57_143 | 21:02 |
| bauwser | lascii: and I wonder if because of I293da6f320dd8f3474ce2a9c907298e1fb348181 it's not just going into https://github.com/openstack/nova/blob/master/nova/cells/messaging.py#L1041 | 21:03 |
| *** sachin has quit IRC | 21:03 | |
| *** krtaylor has joined #openstack-nova | 21:03 | |
| bauwser | lascii: because the req-ids seem to match with requests to delete an instance | 21:03 |
| bauwser | lascii: or negative checks | 21:04 |
| *** sachin has joined #openstack-nova | 21:04 | |
| *** aysyd has quit IRC | 21:05 | |
| *** sachin has quit IRC | 21:06 | |
| *** sachin has joined #openstack-nova | 21:06 | |
| bauwser | lascii: by looking at the compare and swap change, it was previously raising a NotFound, while now it's a InstanceUpdateConflict | 21:07 |
| lascii | bauwser: what I'm not sure of yet is if the message is logged once because it was tried once and then succeeded, or because it retried multiple times and still failed | 21:07 |
| bauwser | lascii: I don't think so | 21:08 |
| bauwser | lascii: any bug I could attach my change ? | 21:08 |
| bauwser | lascii: the change is pretty straight | 21:08 |
| lascii | bauwser: I am not aware of an open bug. I thought you might have opened one though | 21:09 |
| bauwser | lascii: nope, will do then | 21:09 |
| bauwser | lascii: that's not technically an issue, rather a log | 21:09 |
| bauwser | lascii: but opening a bug for tracing that | 21:09 |
| *** dustins has quit IRC | 21:10 | |
| *** dustins has joined #openstack-nova | 21:10 | |
| *** dustins has quit IRC | 21:11 | |
| *** hemna has quit IRC | 21:11 | |
| lascii | what concerns me is that when that exception is raised it's supposed to retry up to 5 times. so if it's getting raised to cells code that means there were 6 failures | 21:11 |
| bauwser | lascii: but we were catching an InstanceNotFound right? | 21:12 |
| bauwser | for the rechecks | 21:12 |
| lascii | bauwser: I'm looking at _retry_instance_update in https://review.openstack.org/#/c/202593/4/nova/db/sqlalchemy/api.py,cm | 21:13 |
| bauwser | ah that | 21:13 |
| bauwser | lascii: but whatever, if the instance is not found, it can retry X times, it will not eventually find it | 21:14 |
| bauwser | lascii: so, we were having a catch for that | 21:14 |
| bauwser | lascii: (the one I mentioned ie. https://github.com/openstack/nova/blob/master/nova/cells/messaging.py#L1041) | 21:14 |
| lascii | right, that doesn't get retried | 21:15 |
| *** apuimedo has quit IRC | 21:15 | |
| lascii | the instanceupdateconflict that's logged. that means it's failing 6 times | 21:15 |
| bauwser | lascii: before the above change, it was calling _instance_get_by_uuid() which was retrying 5 times as well | 21:15 |
| *** apuimedo_ has quit IRC | 21:16 | |
| bauwser | lascii: but at the end, it was still raising a NotFound which was catched by the top-level cell | 21:16 |
| *** burgerk_ has quit IRC | 21:16 | |
| bauwser | lascii: so, here, since we changed the exception, it's now showing in the logs while it was hidden before | 21:17 |
| bauwser | lascii: but I guess that the behaviour was the same, ie. retry 5 times and then raise the exception | 21:17 |
| lascii | I don't think we've changed the exception. I think we added a new one to a case that used to silently succeed(incorrectly) | 21:18 |
| bauwser | lascii: I wasn't involved in the review process for 202593, I need to dig into it | 21:20 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Catch InstanceUpdateConflict when saving Instance at the top cell level https://review.openstack.org/218473 | 21:20 |
| bauwser | lascii: ^ that is a debatable change | 21:20 |
| lascii | bauwser: thanks | 21:21 |
| ndipanov | superdan, responded - pls take a look if you get a chance | 21:21 |
| bauwser | lascii: mmm, reading https://review.openstack.org/#/c/202593/4/nova/db/sqlalchemy/api.py,cm | 21:23 |
| *** mjturek1 has quit IRC | 21:23 | |
| bauwser | lascii: so, now we're using _model_query, right? | 21:23 |
| bauwser | lascii: but then, we use update_on_match() which can raise NoRowsMatched | 21:25 |
| lascii | it always used model_query, it just used to be in _instance_get_by_uuid | 21:26 |
| lascii | yeah, update_on_match is new | 21:26 |
| *** eglynn has quit IRC | 21:27 | |
| bauwser | lascii: in that case, it goes into the exception | 21:27 |
| *** eglynn has joined #openstack-nova | 21:27 | |
| bauwser | lascii: but since we're using instance_update_and_get_original it provides an original arg | 21:27 |
| bauwser | lascii: at the end, it throws a Conflict exc | 21:27 |
| *** eharney has joined #openstack-nova | 21:28 | |
| *** bkopilov has quit IRC | 21:28 | |
| *** tjones has joined #openstack-nova | 21:29 | |
| *** SourabhP has quit IRC | 21:30 | |
| lascii | InstanceUpdateConflict is defined twice in exception.py | 21:31 |
| bauwser | lascii: that's another problem | 21:31 |
| lascii | one has more verbose output than what is in the logs | 21:31 |
| bauwser | lascii: agreed, but it seems we're throwing the right one, because that's the last one :) | 21:32 |
| jaypipes | superdan: no, in the air flying to ATL | 21:33 |
| *** bkopilov has joined #openstack-nova | 21:34 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: Remove duplicate exception https://review.openstack.org/218482 | 21:35 |
| bauwser | lascii: well, your change is good but mdbooth's exception was using 3 parameters | 21:36 |
| bauwser | lascii: I wonder if we should also change the callers | 21:36 |
| bauwser | (that said, I haven't checked the callers) | 21:36 |
| lascii | https://github.com/openstack/nova/blob/master/nova/exception.py#L1445 uses three parameters | 21:36 |
| *** mriedem has quit IRC | 21:37 | |
| bauwser | lascii: sec | 21:37 |
| *** SourabhP has joined #openstack-nova | 21:38 | |
| bauwser | lascii: yeah, we're actually sending an exception that wasn't written by mdbooth :) | 21:38 |
| *** dimsum__ has quit IRC | 21:38 | |
| bauwser | lascii: so the change added one exception with 3 parameters but it was eaten by an already present duplicate which was later in the file and was having only one param | 21:39 |
| lascii | yeah. removing that will give more info in the traceback | 21:39 |
| bauwser | lascii: yup | 21:40 |
| bauwser | lascii: I checked, there were no callers with that one | 21:40 |
| bauwser | lascii: my concern was to have an exception which was possibly set by callers passing only one arg | 21:41 |
| bauwser | lascii: but that's fine | 21:41 |
| bauwser | lascii: anyway, I think the path I followed is correct | 21:41 |
| bauwser | lascii: since we call instance_update_and_get_original(), we're not calling again _instance_get_by_uuid() | 21:42 |
| *** bkopilov has quit IRC | 21:42 | |
| bauwser | lascii: https://review.openstack.org/#/c/202593/4/nova/db/sqlalchemy/api.py,cm L2572 | 21:42 |
| lascii | it'll just get called on a retry | 21:45 |
| *** annegentle has quit IRC | 21:47 | |
| *** jaypipes has quit IRC | 21:47 | |
| *** annegentle has joined #openstack-nova | 21:47 | |
| *** mwagner_lap has joined #openstack-nova | 21:47 | |
| *** meghal has joined #openstack-nova | 21:48 | |
| *** meghal1 has quit IRC | 21:48 | |
| bauwser | lascii: I don't get it | 21:48 |
| *** achanda has joined #openstack-nova | 21:48 | |
| bauwser | lascii: the decorator calls wrap_db_retry which passes a function as a arg to verify the exc type | 21:49 |
| *** jaypipes has joined #openstack-nova | 21:49 | |
| bauwser | lascii: I don't see where _instance_get_by_uuid() is called | 21:50 |
| lascii | line 2468 | 21:50 |
| lascii | but I just realized that it catches UnknownInstanceUpdateConflict, not InstanceUpdateConflict | 21:50 |
| *** hemna has joined #openstack-nova | 21:51 | |
| *** electrocucaracha has quit IRC | 21:52 | |
| lascii | so it's not retrying 6 times like I thought. So I wonder if we should retry in your change. | 21:52 |
| *** eglynn has quit IRC | 21:52 | |
| *** jaypipes has quit IRC | 21:53 | |
| lascii | but I'll leave that for Monday | 21:53 |
| *** lascii is now known as alaski | 21:53 | |
| bauwser | alaski: oh gotcha | 21:53 |
| *** jaypipes has joined #openstack-nova | 21:53 | |
| *** electrocucaracha has joined #openstack-nova | 21:53 | |
| bauwser | alaski: sure have a good week-end | 21:53 |
| *** ctrath has quit IRC | 21:53 | |
| alaski | thanks, you too | 21:54 |
| openstackgerrit | jichenjc proposed openstack/nova: Remove duplicate VALID_NAME_REGEX https://review.openstack.org/218489 | 21:54 |
| *** SourabhP has quit IRC | 21:55 | |
| *** SourabhP has joined #openstack-nova | 21:56 | |
| *** boden has quit IRC | 21:56 | |
| *** dimsum__ has joined #openstack-nova | 21:56 | |
| jaypipes | bauwser: excellent work on the request spec patch series. all reviewed now. | 21:58 |
| *** Apoorva has quit IRC | 21:58 | |
| *** tjones has quit IRC | 21:59 | |
| bauwser | jaypipes: cool, thanks for your support | 21:59 |
| *** Apoorva has joined #openstack-nova | 21:59 | |
| bauwser | jaypipes: wanted some way to take a nap during flights ? | 21:59 |
| bauwser | jaypipes: my series seems a good opportunity for that :) | 21:59 |
| *** dims_ has joined #openstack-nova | 22:00 | |
| bauwser | jaypipes: but I prefer taking melatonin instead :) | 22:00 |
| *** xyang1 has quit IRC | 22:00 | |
| bauwser | well, Friday is off | 22:00 |
| *** bauwser is now known as bauzas | 22:00 | |
| *** haomaiwang has quit IRC | 22:01 | |
| openstackgerrit | jichenjc proposed openstack/nova: devref: change error messages no need microversion https://review.openstack.org/218491 | 22:01 |
| *** SourabhP has quit IRC | 22:01 | |
| *** bkopilov has joined #openstack-nova | 22:01 | |
| bauzas | g'night folks | 22:01 |
| ndipanov | bye | 22:02 |
| *** haomaiwang has joined #openstack-nova | 22:02 | |
| *** dimsum__ has quit IRC | 22:02 | |
| *** jichen has quit IRC | 22:04 | |
| *** mdrabe has quit IRC | 22:05 | |
| *** SourabhP has joined #openstack-nova | 22:05 | |
| *** tpeoples has quit IRC | 22:06 | |
| *** tjones has joined #openstack-nova | 22:06 | |
| *** thorst1 has quit IRC | 22:08 | |
| jaypipes | bauzas: :) | 22:08 |
| *** jaypipes has quit IRC | 22:08 | |
| *** SourabhP has quit IRC | 22:08 | |
| *** tjones has left #openstack-nova | 22:09 | |
| *** meghal1 has joined #openstack-nova | 22:09 | |
| *** meghal has quit IRC | 22:11 | |
| *** edmondsw has quit IRC | 22:17 | |
| *** Apoorva has quit IRC | 22:24 | |
| *** yixing has joined #openstack-nova | 22:24 | |
| *** Apoorva has joined #openstack-nova | 22:24 | |
| *** bnemec has quit IRC | 22:27 | |
| anish | hemna: iser is supported | 22:27 |
| anish | iser doesn't required a special name | 22:27 |
| *** zzzeek has quit IRC | 22:27 | |
| hemna | anish, yes I know, but the member called supported_transports doesn't have it. | 22:28 |
| hemna | so it's confusing and misleading | 22:28 |
| hemna | this stuff needs to be maintainable by more than you :) | 22:28 |
| *** emagana has joined #openstack-nova | 22:28 | |
| anish | well iser is a special case of iface as well as transport | 22:28 |
| anish | but okay | 22:28 |
| anish | lemme add it in | 22:28 |
| hemna | and/or document it in the code why | 22:28 |
| hemna | thanks | 22:29 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: compute: migrate/resize paths properly handle stashed numa_topology https://review.openstack.org/218500 | 22:29 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: compute: migrate/resize paths properly handle stashed numa_topology https://review.openstack.org/218500 | 22:35 |
| anish | hemna: done and done | 22:36 |
| anish | whoops, wrong channel | 22:36 |
| *** hemna is now known as hemnafk | 22:39 | |
| *** ndipanov has quit IRC | 22:42 | |
| *** eharney has quit IRC | 22:47 | |
| *** tonytan4ever has quit IRC | 22:55 | |
| *** rook has quit IRC | 22:57 | |
| *** rook has joined #openstack-nova | 22:58 | |
| *** yamahata has joined #openstack-nova | 22:58 | |
| *** humble__ has joined #openstack-nova | 22:59 | |
| *** yixing has quit IRC | 22:59 | |
| *** humble_ has quit IRC | 23:00 | |
| *** jamielennox is now known as jamielennox|away | 23:00 | |
| *** haomaiwang has quit IRC | 23:01 | |
| *** nagyz_ has joined #openstack-nova | 23:01 | |
| *** haomaiwa_ has joined #openstack-nova | 23:02 | |
| *** nagyz has quit IRC | 23:04 | |
| *** annegentle has quit IRC | 23:04 | |
| *** dannywilson has quit IRC | 23:05 | |
| *** baoli has quit IRC | 23:10 | |
| *** Apoorva has quit IRC | 23:13 | |
| *** sachin has quit IRC | 23:13 | |
| *** sachin has joined #openstack-nova | 23:15 | |
| *** sachin has quit IRC | 23:16 | |
| *** Apoorva has joined #openstack-nova | 23:17 | |
| *** sachin has joined #openstack-nova | 23:17 | |
| *** sachin has quit IRC | 23:19 | |
| *** tonytan4ever has joined #openstack-nova | 23:19 | |
| *** mtanino has quit IRC | 23:21 | |
| *** sachin has joined #openstack-nova | 23:21 | |
| *** david-lyle has quit IRC | 23:24 | |
| *** sachin has quit IRC | 23:26 | |
| *** tjones has joined #openstack-nova | 23:27 | |
| *** vishwanathj has quit IRC | 23:28 | |
| *** sachin has joined #openstack-nova | 23:29 | |
| *** Apoorva has quit IRC | 23:30 | |
| *** sachin has quit IRC | 23:30 | |
| *** sachin has joined #openstack-nova | 23:30 | |
| *** Apoorva has joined #openstack-nova | 23:32 | |
| *** sachin has quit IRC | 23:35 | |
| *** sachin has joined #openstack-nova | 23:36 | |
| *** rook has quit IRC | 23:38 | |
| *** sachin has quit IRC | 23:41 | |
| *** Apoorva has quit IRC | 23:41 | |
| *** sachin has joined #openstack-nova | 23:43 | |
| *** salv-orlando has quit IRC | 23:45 | |
| *** electrocucaracha has quit IRC | 23:46 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/216968 | 23:46 |
| *** rook has joined #openstack-nova | 23:48 | |
| *** emagana has quit IRC | 23:51 | |
| *** emagana has joined #openstack-nova | 23:51 | |
| *** dimsum__ has joined #openstack-nova | 23:52 | |
| *** tjones has quit IRC | 23:54 | |
| *** dims_ has quit IRC | 23:55 | |
| *** emagana has quit IRC | 23:56 | |
| *** tjones has joined #openstack-nova | 23:56 | |
| *** tjones has left #openstack-nova | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
