| *** xchu has joined #openstack-dev | 00:00 | |
| *** markwash has joined #openstack-dev | 00:01 | |
| *** jcooley_ has quit IRC | 00:02 | |
| *** mlavalle has quit IRC | 00:03 | |
| *** dolphm has quit IRC | 00:05 | |
| *** jhesketh__ has joined #openstack-dev | 00:05 | |
| *** kenperkins has joined #openstack-dev | 00:05 | |
| *** jcooley_ has joined #openstack-dev | 00:06 | |
| *** hemna is now known as hemnafk | 00:06 | |
| *** kenperkins has quit IRC | 00:07 | |
| *** jcooley_ has quit IRC | 00:08 | |
| *** tonix has quit IRC | 00:08 | |
| *** nermina has joined #openstack-dev | 00:08 | |
| *** marekd|away has joined #openstack-dev | 00:08 | |
| *** markwash has quit IRC | 00:09 | |
| *** kenperkins has joined #openstack-dev | 00:09 | |
| *** jcooley_ has joined #openstack-dev | 00:10 | |
| *** pmathews has quit IRC | 00:11 | |
| *** kenperkins has quit IRC | 00:11 | |
| *** kenperkins has joined #openstack-dev | 00:12 | |
| *** mikeoutland has joined #openstack-dev | 00:14 | |
| *** DennyZhang has joined #openstack-dev | 00:14 | |
| *** eharney has quit IRC | 00:15 | |
| *** DennyZhang has quit IRC | 00:15 | |
| *** DennyZhang has joined #openstack-dev | 00:16 | |
| *** mriedem has joined #openstack-dev | 00:17 | |
| *** rnirmal has quit IRC | 00:19 | |
| *** teran has quit IRC | 00:20 | |
| *** amotoki has joined #openstack-dev | 00:21 | |
| *** buzztroll has joined #openstack-dev | 00:21 | |
| *** xchu has quit IRC | 00:23 | |
| *** mriedem has quit IRC | 00:23 | |
| *** rfolco has quit IRC | 00:24 | |
| *** teran has joined #openstack-dev | 00:24 | |
| *** dkranz has joined #openstack-dev | 00:24 | |
| *** s0mik has quit IRC | 00:25 | |
| *** mriedem has joined #openstack-dev | 00:25 | |
| *** s0mik has joined #openstack-dev | 00:26 | |
| *** nermina has quit IRC | 00:27 | |
| *** DennyZhang has quit IRC | 00:27 | |
| *** diogogmt has quit IRC | 00:29 | |
| *** mriedem has quit IRC | 00:29 | |
| *** Mandell has quit IRC | 00:31 | |
| *** marun has joined #openstack-dev | 00:31 | |
| *** jamielennox is now known as jamielennox|away | 00:32 | |
| *** Alexei_987 has quit IRC | 00:32 | |
| *** nelsnelson is now known as nelsnelson_away | 00:32 | |
| *** sthaha has joined #openstack-dev | 00:33 | |
| *** sthaha has quit IRC | 00:33 | |
| *** sthaha has joined #openstack-dev | 00:33 | |
| *** yamahata has joined #openstack-dev | 00:33 | |
| *** Mandell has joined #openstack-dev | 00:34 | |
| *** buzztroll has quit IRC | 00:34 | |
| *** nermina has joined #openstack-dev | 00:36 | |
| *** nelsnelson_away has quit IRC | 00:36 | |
| *** dims has quit IRC | 00:37 | |
| *** freyes has joined #openstack-dev | 00:38 | |
| *** faramir1 has joined #openstack-dev | 00:38 | |
| *** dims has joined #openstack-dev | 00:39 | |
| *** comay has quit IRC | 00:40 | |
| bknudson | morganfainberg: did you have a suggestion for a better commit message for https://review.openstack.org/#/c/50488/ ? | 00:40 |
|---|---|---|
| morganfainberg | bknudson, oh, uhm | 00:41 |
| *** markwash has joined #openstack-dev | 00:41 | |
| morganfainberg | bknudson, i think my only concern was the "and" in the commit message | 00:41 |
| morganfainberg | bknudson, it used to be <closes> \n and \n <closes> | 00:42 |
| morganfainberg | bknudson, that was solved. | 00:42 |
| *** morazi has quit IRC | 00:42 | |
| *** michchap has quit IRC | 00:42 | |
| morganfainberg | i'll +2/+A it. | 00:42 |
| bknudson | morganfainberg: ok, thanks. that one has had plenty of reviews. | 00:42 |
| morganfainberg | bknudson, actually... my browser isn't loading review.openstack.org atm | 00:42 |
| *** michchap has joined #openstack-dev | 00:42 | |
| morganfainberg | bknudson, let me see if i can figure it out. if not, you can reference this irc chat and +2/+A :) | 00:43 |
| morganfainberg | bknudson, there we og. dns was ... lame | 00:43 |
| *** comay has joined #openstack-dev | 00:46 | |
| *** freyes has quit IRC | 00:47 | |
| *** dims has quit IRC | 00:47 | |
| *** xqueralt has quit IRC | 00:48 | |
| spandhe | using nova-api, is there a way to get information about deleted VMs for a tenant/user? | 00:48 |
| spandhe | I actually want to see quota usage from the api, but seems that will be available in havana | 00:49 |
| *** sthaha has quit IRC | 00:49 | |
| *** pixelb has joined #openstack-dev | 00:49 | |
| *** sthaha has joined #openstack-dev | 00:49 | |
| *** novas0x2a|laptop has quit IRC | 00:50 | |
| *** jcooley_ has quit IRC | 00:50 | |
| *** teran_ has joined #openstack-dev | 00:50 | |
| *** mriedem has joined #openstack-dev | 00:52 | |
| *** jcooley_ has joined #openstack-dev | 00:53 | |
| *** teran has quit IRC | 00:53 | |
| *** JordanRinke_afk is now known as JordanRinke | 00:54 | |
| *** nosnos has joined #openstack-dev | 00:54 | |
| *** kenperkins has quit IRC | 00:57 | |
| *** sarob has joined #openstack-dev | 00:57 | |
| *** dims has joined #openstack-dev | 01:02 | |
| *** nelsnelson has joined #openstack-dev | 01:03 | |
| *** e0ne has quit IRC | 01:04 | |
| *** mrodden has quit IRC | 01:04 | |
| *** nosnos has quit IRC | 01:06 | |
| *** nosnos has joined #openstack-dev | 01:06 | |
| *** sarob has quit IRC | 01:07 | |
| *** sarob has joined #openstack-dev | 01:08 | |
| *** dcahill has quit IRC | 01:10 | |
| *** alex_xu has joined #openstack-dev | 01:10 | |
| *** sarob has quit IRC | 01:12 | |
| *** spandhe has quit IRC | 01:13 | |
| *** mohits has quit IRC | 01:13 | |
| *** jsavak has quit IRC | 01:14 | |
| *** stevemar has joined #openstack-dev | 01:15 | |
| *** jamielennox|away is now known as jamielennox | 01:15 | |
| *** freyes has joined #openstack-dev | 01:16 | |
| *** bingbu has joined #openstack-dev | 01:17 | |
| *** sarob has joined #openstack-dev | 01:17 | |
| *** pixelb has quit IRC | 01:21 | |
| *** nermina has quit IRC | 01:22 | |
| *** alop has quit IRC | 01:26 | |
| *** bingbu has quit IRC | 01:26 | |
| *** bswartz has quit IRC | 01:28 | |
| *** bswartz has joined #openstack-dev | 01:30 | |
| *** rods has quit IRC | 01:30 | |
| *** xarses has quit IRC | 01:34 | |
| *** bdpayne has quit IRC | 01:34 | |
| *** buzztroll has joined #openstack-dev | 01:35 | |
| *** amcrn has quit IRC | 01:37 | |
| *** colinmcnamara has quit IRC | 01:37 | |
| *** buzztroll has quit IRC | 01:39 | |
| *** buzztroll has joined #openstack-dev | 01:40 | |
| *** lbragstad has quit IRC | 01:40 | |
| *** yaguang has joined #openstack-dev | 01:40 | |
| *** ljjjustin has joined #openstack-dev | 01:41 | |
| *** joesavak has joined #openstack-dev | 01:42 | |
| *** bingbu has joined #openstack-dev | 01:42 | |
| *** lbragstad has joined #openstack-dev | 01:43 | |
| *** nermina has joined #openstack-dev | 01:43 | |
| *** jsavak has joined #openstack-dev | 01:43 | |
| *** ljjjustin has quit IRC | 01:46 | |
| *** joesavak has quit IRC | 01:47 | |
| *** epim has quit IRC | 01:47 | |
| *** melwitt has quit IRC | 01:48 | |
| *** ljjjustin has joined #openstack-dev | 01:48 | |
| *** s0mik has quit IRC | 01:50 | |
| *** shakayumi has joined #openstack-dev | 01:51 | |
| *** jergerber has quit IRC | 01:53 | |
| *** mrodden has joined #openstack-dev | 01:53 | |
| *** nermina has quit IRC | 01:54 | |
| *** buzztroll has quit IRC | 01:54 | |
| *** akrivoka has quit IRC | 01:55 | |
| *** mrodden1 has joined #openstack-dev | 01:55 | |
| *** tanisdl has quit IRC | 01:55 | |
| *** jayg is now known as jayg|g0n3 | 01:56 | |
| *** sushils has quit IRC | 01:57 | |
| *** mrodden has quit IRC | 01:57 | |
| *** radix_ has quit IRC | 01:58 | |
| *** nati_ueno has quit IRC | 02:01 | |
| *** dolphm has joined #openstack-dev | 02:04 | |
| *** agntdrake has quit IRC | 02:04 | |
| *** angdraug has quit IRC | 02:05 | |
| *** jcooley_ has quit IRC | 02:08 | |
| *** s0mik has joined #openstack-dev | 02:10 | |
| *** dave_tucker has quit IRC | 02:11 | |
| *** s0mik has quit IRC | 02:14 | |
| *** teran has joined #openstack-dev | 02:14 | |
| *** e0ne has joined #openstack-dev | 02:15 | |
| *** markvoelker1 has joined #openstack-dev | 02:17 | |
| stevemar | ayoung: filling in extension_data in an extension data, what should we put for namespace and updated? | 02:17 |
| ayoung | stevemar, something that someone would have to think really hard about, so that it slips unnoticed through the code review | 02:17 |
| *** teran_ has quit IRC | 02:18 | |
| stevemar | :) | 02:18 |
| ayoung | usually namespace is the extension name | 02:18 |
| ayoung | updated is the time the extension was updated, but we probably haven't been keeping those up to date | 02:18 |
| *** erkules_ has joined #openstack-dev | 02:18 | |
| ayoung | just put the current time in there | 02:18 |
| *** buzztroll has joined #openstack-dev | 02:19 | |
| *** bingbu has quit IRC | 02:19 | |
| *** e0ne has quit IRC | 02:20 | |
| *** devoid has joined #openstack-dev | 02:21 | |
| *** erkules has quit IRC | 02:21 | |
| *** xarses has joined #openstack-dev | 02:22 | |
| *** xmltok_ has quit IRC | 02:23 | |
| stevemar | ayoung: m'alright | 02:24 |
| *** nermina has joined #openstack-dev | 02:24 | |
| *** markwash has quit IRC | 02:24 | |
| *** bingbu has joined #openstack-dev | 02:25 | |
| *** xjiujiu has joined #openstack-dev | 02:28 | |
| *** giroro_ has quit IRC | 02:28 | |
| *** phschwartz is now known as phschwartz_aw | 02:28 | |
| *** devoid has quit IRC | 02:29 | |
| *** nati_ueno has joined #openstack-dev | 02:30 | |
| *** nermina has quit IRC | 02:31 | |
| *** faramir1 has quit IRC | 02:31 | |
| *** Mandell has quit IRC | 02:32 | |
| *** Ruetobas has joined #openstack-dev | 02:35 | |
| *** xjiujiu has quit IRC | 02:35 | |
| *** xjiujiu has joined #openstack-dev | 02:35 | |
| *** Ruetobas has quit IRC | 02:39 | |
| *** Ruetobas has joined #openstack-dev | 02:40 | |
| *** jp_at_hp1 has quit IRC | 02:40 | |
| *** jsavak has quit IRC | 02:42 | |
| *** reed has quit IRC | 02:42 | |
| *** ayoung has quit IRC | 02:44 | |
| *** xjiujiu has quit IRC | 02:44 | |
| *** xjiujiu has joined #openstack-dev | 02:44 | |
| *** xjiujiu has joined #openstack-dev | 02:46 | |
| *** vkozhukalov has joined #openstack-dev | 02:48 | |
| *** Ruetobas has quit IRC | 02:48 | |
| *** Ruetobas has joined #openstack-dev | 02:48 | |
| *** freyes has quit IRC | 02:49 | |
| *** jimjiang has quit IRC | 02:51 | |
| *** faramir1 has joined #openstack-dev | 02:51 | |
| *** jimjiang has joined #openstack-dev | 02:52 | |
| *** dims has quit IRC | 02:53 | |
| *** dims has joined #openstack-dev | 02:53 | |
| *** renlt has joined #openstack-dev | 02:53 | |
| *** mikeoutland has quit IRC | 02:55 | |
| *** jimjiang has quit IRC | 02:57 | |
| *** _dbalog has joined #openstack-dev | 02:57 | |
| *** jimjiang has joined #openstack-dev | 02:58 | |
| *** giulivo has quit IRC | 03:00 | |
| *** ctracey is now known as ctracey|away | 03:01 | |
| *** nermina has joined #openstack-dev | 03:04 | |
| *** xjiujiu has quit IRC | 03:05 | |
| *** thomasem has joined #openstack-dev | 03:06 | |
| *** yamahata__ has quit IRC | 03:06 | |
| *** yamahata__ has joined #openstack-dev | 03:06 | |
| *** thomasem has quit IRC | 03:12 | |
| *** paragan has joined #openstack-dev | 03:14 | |
| *** paragan has quit IRC | 03:14 | |
| *** paragan has joined #openstack-dev | 03:14 | |
| *** faramir1 has quit IRC | 03:16 | |
| *** buzztroll has quit IRC | 03:18 | |
| *** ljjjustin has quit IRC | 03:21 | |
| *** rushiagr has joined #openstack-dev | 03:23 | |
| *** shakayumi has quit IRC | 03:24 | |
| *** dolphm has quit IRC | 03:25 | |
| *** faramir1 has joined #openstack-dev | 03:29 | |
| *** dolphm has joined #openstack-dev | 03:30 | |
| *** rcleere has joined #openstack-dev | 03:35 | |
| *** otherwiseguy has quit IRC | 03:40 | |
| *** rongze has joined #openstack-dev | 03:40 | |
| *** maxskew has joined #openstack-dev | 03:45 | |
| *** arnaud has quit IRC | 03:51 | |
| *** arnaud__ has quit IRC | 03:51 | |
| *** bingbu has quit IRC | 03:51 | |
| *** rraja has joined #openstack-dev | 03:52 | |
| *** buzztroll has joined #openstack-dev | 03:53 | |
| *** aveiga has quit IRC | 03:53 | |
| *** jecarey has joined #openstack-dev | 03:54 | |
| *** jimjiang has quit IRC | 03:55 | |
| *** mriedem has quit IRC | 03:55 | |
| *** jimjiang has joined #openstack-dev | 03:56 | |
| *** aditirav has joined #openstack-dev | 03:57 | |
| *** aditirav has quit IRC | 03:58 | |
| *** aditirav has joined #openstack-dev | 03:58 | |
| *** renlt has quit IRC | 03:59 | |
| *** renlt has joined #openstack-dev | 03:59 | |
| *** mdomsch has joined #openstack-dev | 04:01 | |
| *** bingbu has joined #openstack-dev | 04:04 | |
| *** dolphm has quit IRC | 04:05 | |
| *** mikeoutland has joined #openstack-dev | 04:07 | |
| *** maxskew has left #openstack-dev | 04:11 | |
| *** buzztroll has quit IRC | 04:12 | |
| *** itooon has joined #openstack-dev | 04:14 | |
| *** chandankumar has joined #openstack-dev | 04:16 | |
| *** DennyZhang has joined #openstack-dev | 04:18 | |
| *** DennyZhang has quit IRC | 04:21 | |
| *** DennyZhang has joined #openstack-dev | 04:22 | |
| *** DennyZhang has quit IRC | 04:23 | |
| *** DennyZhang has joined #openstack-dev | 04:24 | |
| *** pmathews has joined #openstack-dev | 04:24 | |
| *** jimjiang has quit IRC | 04:25 | |
| *** eglynn_ has quit IRC | 04:25 | |
| *** jimjiang has joined #openstack-dev | 04:26 | |
| *** buzztroll has joined #openstack-dev | 04:26 | |
| *** pmathews has quit IRC | 04:28 | |
| *** DennyZhang has quit IRC | 04:29 | |
| *** aeperezt has quit IRC | 04:31 | |
| *** rongze has quit IRC | 04:33 | |
| *** epopt37 has quit IRC | 04:34 | |
| *** DennyZhang has joined #openstack-dev | 04:38 | |
| *** buzztroll has quit IRC | 04:42 | |
| *** pmathews has joined #openstack-dev | 04:43 | |
| *** buzztroll has joined #openstack-dev | 04:46 | |
| *** epopt37 has joined #openstack-dev | 04:46 | |
| *** pdevine has joined #openstack-dev | 04:49 | |
| *** Mandell has joined #openstack-dev | 04:49 | |
| *** SergeyLukjanov has joined #openstack-dev | 04:49 | |
| *** rongze has joined #openstack-dev | 04:55 | |
| *** sarob_ has joined #openstack-dev | 04:56 | |
| *** sarob has quit IRC | 04:58 | |
| *** rongze_ has joined #openstack-dev | 04:59 | |
| *** jecarey has quit IRC | 04:59 | |
| *** rongze has quit IRC | 05:00 | |
| *** sarob_ has quit IRC | 05:01 | |
| *** aditirav_ has joined #openstack-dev | 05:03 | |
| *** rushiagr has quit IRC | 05:04 | |
| *** rongze_ has quit IRC | 05:04 | |
| *** qs201 has joined #openstack-dev | 05:05 | |
| *** qs201 has quit IRC | 05:07 | |
| *** aditirav has quit IRC | 05:07 | |
| *** aditirav_ has quit IRC | 05:07 | |
| *** DennyZhang has quit IRC | 05:09 | |
| *** aditirav has joined #openstack-dev | 05:09 | |
| *** tserong has quit IRC | 05:13 | |
| *** teran has quit IRC | 05:13 | |
| *** rraja has quit IRC | 05:14 | |
| *** rektide has quit IRC | 05:17 | |
| *** zaitcev has quit IRC | 05:20 | |
| *** stevemar has quit IRC | 05:20 | |
| *** reed has joined #openstack-dev | 05:21 | |
| *** guohliu has joined #openstack-dev | 05:23 | |
| *** sandywalsh_ has quit IRC | 05:23 | |
| *** achampion has quit IRC | 05:23 | |
| *** achampion has joined #openstack-dev | 05:24 | |
| *** alop has joined #openstack-dev | 05:26 | |
| *** reed has quit IRC | 05:26 | |
| *** cdub_ has quit IRC | 05:29 | |
| *** neeti has joined #openstack-dev | 05:29 | |
| *** alop has quit IRC | 05:30 | |
| *** alop has joined #openstack-dev | 05:32 | |
| *** spzala_ has quit IRC | 05:32 | |
| *** cpallares has quit IRC | 05:33 | |
| *** sgordon has quit IRC | 05:33 | |
| *** achampion has quit IRC | 05:33 | |
| *** achampion has joined #openstack-dev | 05:33 | |
| *** haomaiwang has joined #openstack-dev | 05:34 | |
| *** cdub has joined #openstack-dev | 05:35 | |
| *** buzztroll has quit IRC | 05:35 | |
| *** garyk has quit IRC | 05:35 | |
| *** bingbu has quit IRC | 05:36 | |
| *** sandywalsh_ has joined #openstack-dev | 05:39 | |
| *** Ryan_Lane has quit IRC | 05:41 | |
| *** Ryan_Lane has joined #openstack-dev | 05:42 | |
| *** prekarat has joined #openstack-dev | 05:44 | |
| *** Ryan_Lane has quit IRC | 05:46 | |
| *** markwash has joined #openstack-dev | 05:47 | |
| *** rohitk has joined #openstack-dev | 05:48 | |
| *** bingbu has joined #openstack-dev | 05:49 | |
| *** ffio has quit IRC | 05:50 | |
| *** jhesketh__ has quit IRC | 05:50 | |
| *** jhesketh__ has joined #openstack-dev | 05:51 | |
| *** sgordon has joined #openstack-dev | 05:52 | |
| *** aditirav has quit IRC | 05:56 | |
| *** aditirav_ has joined #openstack-dev | 05:56 | |
| *** erkules_ has quit IRC | 05:56 | |
| *** aditirav_ has quit IRC | 05:57 | |
| *** swaT30 has quit IRC | 05:58 | |
| *** aditirav has joined #openstack-dev | 05:58 | |
| *** rushiagr has joined #openstack-dev | 05:58 | |
| *** erkules_ has joined #openstack-dev | 05:59 | |
| *** ffio has joined #openstack-dev | 05:59 | |
| *** swaT30 has joined #openstack-dev | 06:00 | |
| *** itooon_ has joined #openstack-dev | 06:00 | |
| *** aditirav_ has joined #openstack-dev | 06:02 | |
| *** itooon has quit IRC | 06:02 | |
| *** itooon_ is now known as itooon | 06:02 | |
| *** erkules_ is now known as erkules | 06:04 | |
| *** aditirav has quit IRC | 06:04 | |
| *** aditirav_ is now known as aditirav | 06:04 | |
| *** edleafe- has quit IRC | 06:06 | |
| *** edleafe has joined #openstack-dev | 06:06 | |
| *** Hien has quit IRC | 06:07 | |
| *** Hien has joined #openstack-dev | 06:07 | |
| *** aditirav has quit IRC | 06:07 | |
| *** kenperkins has joined #openstack-dev | 06:08 | |
| *** aditirav has joined #openstack-dev | 06:08 | |
| *** adalbas has quit IRC | 06:12 | |
| *** bdpayne has joined #openstack-dev | 06:12 | |
| *** teran has joined #openstack-dev | 06:14 | |
| *** rongze has joined #openstack-dev | 06:15 | |
| *** sgordon has quit IRC | 06:17 | |
| *** rraja has joined #openstack-dev | 06:18 | |
| *** DennyZhang has joined #openstack-dev | 06:19 | |
| *** teran has quit IRC | 06:19 | |
| *** aditirav_ has joined #openstack-dev | 06:23 | |
| *** aditirav has quit IRC | 06:23 | |
| *** aditirav_ is now known as aditirav | 06:24 | |
| *** finchd has joined #openstack-dev | 06:26 | |
| *** DennyZhang has quit IRC | 06:27 | |
| *** DennyZhang has joined #openstack-dev | 06:28 | |
| *** avishay has joined #openstack-dev | 06:28 | |
| *** adalbas has joined #openstack-dev | 06:29 | |
| *** DennyZhang has left #openstack-dev | 06:30 | |
| *** aditirav has quit IRC | 06:33 | |
| *** djinni_ has quit IRC | 06:34 | |
| *** yeylon__ has joined #openstack-dev | 06:34 | |
| *** alex_klimov has joined #openstack-dev | 06:34 | |
| *** rushiagr has quit IRC | 06:36 | |
| *** sgordon has joined #openstack-dev | 06:36 | |
| *** boris-42 has joined #openstack-dev | 06:36 | |
| *** markvoelker1 has quit IRC | 06:36 | |
| *** djinni has joined #openstack-dev | 06:36 | |
| *** markwash has quit IRC | 06:37 | |
| *** alop has quit IRC | 06:39 | |
| *** alex_klimov has quit IRC | 06:39 | |
| *** sthaha has quit IRC | 06:40 | |
| *** markwash has joined #openstack-dev | 06:40 | |
| *** alex_klimov has joined #openstack-dev | 06:41 | |
| *** bdpayne has quit IRC | 06:42 | |
| *** harlowja is now known as harlowja_away | 06:43 | |
| *** HenryG has quit IRC | 06:43 | |
| *** rushiagr has joined #openstack-dev | 06:43 | |
| *** tonyfy has joined #openstack-dev | 06:45 | |
| *** SergeyLukjanov has quit IRC | 06:45 | |
| *** buzztroll has joined #openstack-dev | 06:46 | |
| *** nshaikh has joined #openstack-dev | 06:48 | |
| *** marun has quit IRC | 06:49 | |
| *** buzztroll has quit IRC | 06:51 | |
| *** DennyZhang has joined #openstack-dev | 06:51 | |
| *** vkozhukalov has quit IRC | 06:51 | |
| *** DennyZhang has left #openstack-dev | 06:54 | |
| *** ngoracke1 has quit IRC | 06:54 | |
| *** pdevine has quit IRC | 06:55 | |
| *** DennyZhang has joined #openstack-dev | 06:56 | |
| *** garyk has joined #openstack-dev | 06:56 | |
| *** vartom11111119 has joined #openstack-dev | 06:58 | |
| *** bingbu has quit IRC | 07:00 | |
| *** sgordon has quit IRC | 07:01 | |
| *** michchap has quit IRC | 07:02 | |
| *** kenperkins has quit IRC | 07:02 | |
| *** michchap has joined #openstack-dev | 07:03 | |
| *** neelashah1 has quit IRC | 07:03 | |
| *** jcooley_ has joined #openstack-dev | 07:03 | |
| *** denis_makogon has joined #openstack-dev | 07:05 | |
| *** markwash has quit IRC | 07:06 | |
| *** Abhishek_ has joined #openstack-dev | 07:07 | |
| *** jcooley_ has quit IRC | 07:07 | |
| *** jcooley_ has joined #openstack-dev | 07:09 | |
| *** pmathews has quit IRC | 07:10 | |
| *** HenryG has joined #openstack-dev | 07:10 | |
| *** rushiagr has quit IRC | 07:11 | |
| *** rushiagr has joined #openstack-dev | 07:11 | |
| *** jamielennox is now known as jamielennox|away | 07:11 | |
| *** bingbu has joined #openstack-dev | 07:12 | |
| *** jcooley_ has quit IRC | 07:14 | |
| *** aditirav has joined #openstack-dev | 07:14 | |
| *** yolanda has joined #openstack-dev | 07:15 | |
| *** tserong has joined #openstack-dev | 07:16 | |
| *** mikeoutland has quit IRC | 07:20 | |
| *** jhesketh__ has quit IRC | 07:20 | |
| *** boris-42 has quit IRC | 07:21 | |
| *** DennyZhang has quit IRC | 07:22 | |
| *** doron_afk has joined #openstack-dev | 07:29 | |
| *** Abhishek_ has quit IRC | 07:30 | |
| *** vartom11111119 has quit IRC | 07:31 | |
| *** vartom111111110 has joined #openstack-dev | 07:31 | |
| *** nelsnelson has quit IRC | 07:31 | |
| *** nelsnelson has joined #openstack-dev | 07:32 | |
| *** urulama has joined #openstack-dev | 07:33 | |
| *** zul has quit IRC | 07:34 | |
| *** rohitk has quit IRC | 07:35 | |
| *** DennyZhang has joined #openstack-dev | 07:35 | |
| *** ig has joined #openstack-dev | 07:37 | |
| *** afazekas has joined #openstack-dev | 07:37 | |
| *** DennyZhang has quit IRC | 07:40 | |
| *** evgenyf has joined #openstack-dev | 07:41 | |
| *** lsmola_ has joined #openstack-dev | 07:44 | |
| *** flaper87|afk is now known as flaper87 | 07:46 | |
| *** jimjiang has quit IRC | 07:47 | |
| *** jimjiang has joined #openstack-dev | 07:48 | |
| *** ig1 has joined #openstack-dev | 07:49 | |
| *** ig1 has quit IRC | 07:50 | |
| *** ig has quit IRC | 07:50 | |
| *** ig has joined #openstack-dev | 07:50 | |
| *** julienvey_ has joined #openstack-dev | 07:54 | |
| *** venum has joined #openstack-dev | 07:58 | |
| *** mrunge has joined #openstack-dev | 08:00 | |
| *** teran has joined #openstack-dev | 08:02 | |
| *** arozumenko has joined #openstack-dev | 08:02 | |
| *** kbrierly has quit IRC | 08:03 | |
| *** kdbrierly has quit IRC | 08:08 | |
| *** aditirav has quit IRC | 08:08 | |
| *** amotoki has quit IRC | 08:09 | |
| *** aditirav has joined #openstack-dev | 08:09 | |
| *** xga_ has joined #openstack-dev | 08:10 | |
| *** xga has quit IRC | 08:10 | |
| *** xga_ has quit IRC | 08:13 | |
| *** xga has joined #openstack-dev | 08:13 | |
| ttx | notmyname: o/ | 08:14 |
| notmyname | howdy :-) | 08:14 |
| *** e0ne has joined #openstack-dev | 08:14 | |
| notmyname | brief update for swifty things: python-swiftclient is nearing a 2.0 release | 08:14 |
| notmyname | when https://review.openstack.org/#/c/33473/ lands, both the final 1.x and 2.0 will be tagged | 08:14 |
| notmyname | the 2.0 is important because it adds cert checking | 08:15 |
| notmyname | on by default | 08:15 |
| ttx | notmyname: heh, yes | 08:15 |
| notmyname | and this may break existing clients | 08:15 |
| notmyname | so I'll also tag a final 1.x to get everything that's been merged | 08:15 |
| notmyname | before that | 08:15 |
| ttx | sounds good | 08:15 |
| ttx | that last review seems in good shape | 08:16 |
| ttx | so hopefully will happen before the holiday season | 08:16 |
| notmyname | not really. looks like it may have an issue. just overheard another review of it in -swift | 08:16 |
| *** moted has quit IRC | 08:16 | |
| notmyname | but yes, I hope so | 08:16 |
| notmyname | I'm hoping for tomorrow, actually | 08:17 |
| ttx | it's all +1 now but maybe there is more to it | 08:17 |
| ttx | anything you wanted to see discussed at the meeting ? | 08:17 |
| notmyname | but that brings us to the major issue of the day: gate checks | 08:17 |
| *** tkammer has joined #openstack-dev | 08:17 | |
| notmyname | yes. gate checks | 08:17 |
| notmyname | first, have you heard of other groups in openstack having issues getting things through the gate? | 08:17 |
| ttx | OK, adding to the agenda | 08:18 |
| * ttx was fearing the agenda was a bit light but it suddenly got heavier :) | 08:18 | |
| ttx | notmyname: yes, most people got bitten by the sphinx issue last week | 08:18 |
| ttx | it delayed 2013.2.1 from Thursday to Today | 08:18 |
| notmyname | the sphinx issue was a big deal, but a nice one-off. the problem that I see is that it wasn't really that unusual from a gate status perspective | 08:19 |
| notmyname | ttx: anything beyond that? what we're seeing is that approved patches take _days_ to land | 08:19 |
| notmyname | with many reverify comments | 08:19 |
| notmyname | and groking the logs to do the right bugs is a major time sink | 08:20 |
| notmyname | eg right now a perfectly good patch only has ~70% chance of passing http://not.mn/gate_status.html | 08:20 |
| ttx | notmyname: my understanding was that it was mostly the time to process the backlog, but maybe the non-deterministic issues got busier yesterday | 08:20 |
| notmyname | which is .2% chance for the current gat eto clear | 08:21 |
| ttx | notmyname: let's discuss that at meeting later today | 08:21 |
| *** sergmelikyan has joined #openstack-dev | 08:21 | |
| ttx | notmyname: pretty sure sgaue will have a clearer status than I have | 08:21 |
| notmyname | ttx: please read the text on my graph page to prep | 08:21 |
| ttx | sdague* | 08:21 |
| ttx | notmyname: will do andI'll link it on the agenda, too | 08:21 |
| notmyname | thanks | 08:21 |
| notmyname | ttx: so what I want is 3 things | 08:22 |
| notmyname | 1) know if this is a systemic thing (which I suspect it is) affecting all projects | 08:22 |
| notmyname | 2) find a source rather than just the knowledge of something being wrong | 08:22 |
| notmyname | 3) find a way forward. this must be more than "we'll pay more attention to bugs in the hand-curated elastic-recheck list" | 08:23 |
| notmyname | things like: can we redo the way gate queues are managed? | 08:23 |
| notmyname | do we need to reconsider the way integration tests are done? | 08:23 |
| notmyname | do we need to reconsider the dev methodology? | 08:24 |
| ttx | notmyname: ok, good topic, will try to get the right people around | 08:24 |
| notmyname | ttx: anyway, it's a huge topic, and IMO a huge problem. one that is causing a _lot_ of frustration, at least on our side of things | 08:24 |
| notmyname | and I'm scared that if it continues the consequences may be dire | 08:25 |
| ttx | notmyname: ack | 08:25 |
| *** AMike has quit IRC | 08:25 | |
| *** pfreund has joined #openstack-dev | 08:26 | |
| ttx | notmyname: ok, then see you at the meeting later | 08:26 |
| notmyname | ttx: k, thanks | 08:26 |
| *** vkozhukalov has joined #openstack-dev | 08:26 | |
| *** denis_makogon has quit IRC | 08:26 | |
| *** jtomasek has joined #openstack-dev | 08:28 | |
| *** avishayb has joined #openstack-dev | 08:29 | |
| *** teran has quit IRC | 08:30 | |
| *** noorul has quit IRC | 08:33 | |
| *** gszasz has joined #openstack-dev | 08:34 | |
| *** zul has joined #openstack-dev | 08:34 | |
| *** jcoufal has joined #openstack-dev | 08:34 | |
| *** jhesketh_ has quit IRC | 08:34 | |
| *** DennyZhang has joined #openstack-dev | 08:36 | |
| *** marios has quit IRC | 08:37 | |
| *** DinaBelova has joined #openstack-dev | 08:37 | |
| *** giulivo has joined #openstack-dev | 08:37 | |
| chandankumar | sdague, ping | 08:38 |
| *** marios has joined #openstack-dev | 08:38 | |
| *** kbrierly has joined #openstack-dev | 08:38 | |
| *** rushiagr has quit IRC | 08:38 | |
| *** rushiagr2 has joined #openstack-dev | 08:38 | |
| *** AMike has joined #openstack-dev | 08:38 | |
| *** jimjiang has quit IRC | 08:39 | |
| *** jimjiang has joined #openstack-dev | 08:40 | |
| *** achampion has quit IRC | 08:40 | |
| *** krast has joined #openstack-dev | 08:40 | |
| *** ifarkas has joined #openstack-dev | 08:40 | |
| *** DennyZhang has quit IRC | 08:41 | |
| *** achampion has joined #openstack-dev | 08:41 | |
| *** amuller has joined #openstack-dev | 08:43 | |
| *** HenryG has quit IRC | 08:43 | |
| *** rushiagr2 has quit IRC | 08:44 | |
| *** HenryG has joined #openstack-dev | 08:44 | |
| *** corXi has joined #openstack-dev | 08:45 | |
| *** xingchao has joined #openstack-dev | 08:45 | |
| *** AMike has quit IRC | 08:45 | |
| *** jhesketh_ has joined #openstack-dev | 08:46 | |
| *** AMike has joined #openstack-dev | 08:46 | |
| *** AMike has quit IRC | 08:46 | |
| *** AMike has joined #openstack-dev | 08:46 | |
| *** urulama has quit IRC | 08:47 | |
| *** ygbo has joined #openstack-dev | 08:47 | |
| *** fbo_away is now known as fbo | 08:47 | |
| *** devvesa has joined #openstack-dev | 08:47 | |
| *** _dbalog has quit IRC | 08:48 | |
| *** _dbalog has joined #openstack-dev | 08:48 | |
| *** corXi has quit IRC | 08:49 | |
| *** rushiagr2 has joined #openstack-dev | 08:49 | |
| *** corXi has joined #openstack-dev | 08:50 | |
| *** ogelbukh has joined #openstack-dev | 08:50 | |
| *** jaimegil has joined #openstack-dev | 08:51 | |
| *** asalkeld has quit IRC | 08:51 | |
| *** haypo has joined #openstack-dev | 08:53 | |
| *** asalkeld has joined #openstack-dev | 08:53 | |
| kashyap | chandankumar, Please don't do a 'naked ping', post your topic (ping re: blah). He's in EST time-zone & must be sleeping. | 08:54 |
| *** noorul has joined #openstack-dev | 08:54 | |
| *** urulama has joined #openstack-dev | 08:54 | |
| *** nermina has quit IRC | 08:56 | |
| *** omachace has joined #openstack-dev | 08:56 | |
| chandankumar | kashyap, for this bug, https://bugs.launchpad.net/tempest/+bug/1201201 | 08:58 |
| uvirtbot | Launchpad bug 1201201 in tempest "Document how to install tempest without devstack" [Wishlist,Confirmed] | 08:58 |
| chandankumar | it is telling about which helper script? | 08:58 |
| haypo | hi. i'm trying to update Oslo-incubator in oslo-messaging, but openstack/common/gettextutils.py now depends on a Python module "babel" whereas i don't see this dependency anywhere | 08:58 |
| *** salv-orlando has quit IRC | 09:00 | |
| *** rushiagr2 is now known as rushiagr | 09:00 | |
| *** jistr has joined #openstack-dev | 09:00 | |
| *** vartom111111110 has quit IRC | 09:00 | |
| *** nshaikh has left #openstack-dev | 09:00 | |
| *** ndipanov_gone is now known as ndipanov | 09:01 | |
| *** prekarat has quit IRC | 09:01 | |
| *** boden has joined #openstack-dev | 09:02 | |
| kashyap | chandankumar, IIUC, *you*/bug-asignee has to prepare an install helper script. Hence the bug. | 09:02 |
| *** prekarat has joined #openstack-dev | 09:04 | |
| chandankumar | kashyap, i need more info on helper script, it is related to install tempest through puppet or anisble. | 09:04 |
| chandankumar | ? | 09:04 |
| kashyap | Why don't you try yourself & post the observations on which is optimal. | 09:05 |
| *** xga has quit IRC | 09:05 | |
| *** xga has joined #openstack-dev | 09:06 | |
| *** nosnos has quit IRC | 09:07 | |
| *** nosnos has joined #openstack-dev | 09:07 | |
| *** jpich has joined #openstack-dev | 09:07 | |
| *** derekh has joined #openstack-dev | 09:07 | |
| *** irenab has joined #openstack-dev | 09:07 | |
| *** SergeyLukjanov has joined #openstack-dev | 09:08 | |
| *** SergeyLukjanov has quit IRC | 09:09 | |
| *** yassine has joined #openstack-dev | 09:09 | |
| Adri2000 | am I the only one who would prefer all meeting/event times to be announced in UTC rather than pacific/eastern/... ? :) | 09:09 |
| *** SergeyLukjanov has joined #openstack-dev | 09:09 | |
| StevenK | How about both? | 09:10 |
| *** reidrac has joined #openstack-dev | 09:11 | |
| persia | both is even worse than non-UTC | 09:11 |
| *** mindpixel has joined #openstack-dev | 09:11 | |
| StevenK | Wed 1000EST (Wed 1500Z) is bad? | 09:11 |
| persia | Yes, because 1000EST can also be interpreted as 2400Z, depending on local interpretation | 09:13 |
| chandankumar | kashyap, ok Thanks :) | 09:13 |
| lifeless | persia: uh, but Z adds the disambiguator | 09:13 |
| lifeless | persia: without that you only have multiple local interpretations | 09:13 |
| Adri2000 | then why not western europe time? japan time as well? | 09:13 |
| * persia agrees with Adri2000 | 09:13 | |
| lifeless | persia: so surely both is strictly better than non-UTC alone? | 09:13 |
| *** marekd|away is now known as marekd | 09:14 | |
| lifeless | persia: also, how the hell are you? | 09:14 |
| persia | lifeless: Yes, but UTC alone is better, especially when people get confused about UTC offsets arund the equinox | 09:14 |
| *** omachace has quit IRC | 09:14 | |
| lifeless | persia: did you get to the HK summit? | 09:14 |
| *** omachace has joined #openstack-dev | 09:15 | |
| persia | lifeless: No. I wasn't quite thinking clearly (losses in the family). I'm better now. How are you? | 09:15 |
| *** Ryan_Lane has joined #openstack-dev | 09:15 | |
| *** reidrac has quit IRC | 09:15 | |
| lifeless | persia: oh, thats sad news. I'm great :) | 09:15 |
| lifeless | persia: Cynthia is not growing as fast as the dr's would like, which is a little concerning, but it's early days in the analysis, nothing to panic about [yet] | 09:16 |
| yolanda | hi, i want to create a new feature under oslo, what's the right process to do it? first create a blueprint here? https://blueprints.launchpad.net/oslo/ | 09:17 |
| persia | Oof. May all be well. | 09:17 |
| *** boden has quit IRC | 09:17 | |
| yolanda | it's related to that change: https://review.openstack.org/#/c/61128/ | 09:18 |
| lifeless | interesting | 09:18 |
| lifeless | so I at least am confused about when things move to olso. | 09:18 |
| lifeless | I've been told that they go into the incubator once multiple projects want to use it - but that first they start in one project. | 09:18 |
| lifeless | dhellmann: ^ will be the authority | 09:19 |
| *** johnthetubaguy has joined #openstack-dev | 09:19 | |
| *** yamahata has quit IRC | 09:19 | |
| lifeless | If it's a feature level item, or something needing discussion, then a blueprint is often a useful control point. | 09:19 |
| lifeless | yolanda: was this discussed at the summit? | 09:19 |
| *** Ryan_Lane has quit IRC | 09:20 | |
| yolanda | lifeless, no, we had a blueprint into Ubuntu, and we decided to raise question upstream | 09:20 |
| *** athomas has joined #openstack-dev | 09:20 | |
| *** mrunge has quit IRC | 09:20 | |
| lifeless | yolanda: ok, so I would send an email to the openstack-dev list | 09:20 |
| yolanda | so i just chose one component and keystone, sent a change request to grab opinions about it, and see what's the better way | 09:20 |
| lifeless | yolanda: because this is something that will eventually impact everyone | 09:20 |
| yolanda | lifeless, ok, so i should send an email asking about opinions for it? | 09:21 |
| lifeless | yolanda: (and for instance, I'm extremely puzzled why folk would want to leak this information out). | 09:21 |
| lifeless | yolanda: yeah, definitely. | 09:21 |
| *** mancdaz_away is now known as mancdaz | 09:21 | |
| *** mrunge has joined #openstack-dev | 09:21 | |
| lifeless | I'm fairly sure a bunch of opsec folk will want to understand the benefit of disclosing this operational detail [at all, nevermind in every response] | 09:21 |
| lifeless | yolanda: Then there's the overhead question - why include this in every response; if it's diagnostic material, why not just include it in response to OPTIONS: * | 09:22 |
| persia | yolanda: Do you have a link to the Ubuntu blueprint? What benefit do you expect? | 09:22 |
| yolanda | yes, blueprint is here https://blueprints.launchpad.net/ubuntu/+spec/servercloud-s-server-app-banner-updates | 09:23 |
| yolanda | the intention is for stats collection mostly | 09:23 |
| *** aspiers_ is now known as aspiers | 09:23 | |
| *** safchain has joined #openstack-dev | 09:24 | |
| yolanda | the way i've thought about it, is to include that on config file, and distribution should be statically written in config file, so it's few extra overhead, as call is already parsing this config file anyway | 09:25 |
| yolanda | and as it's included in config file, it can be disabled if people don't want it in their deployment | 09:26 |
| *** eglynn has joined #openstack-dev | 09:27 | |
| *** lucasagomes has joined #openstack-dev | 09:31 | |
| *** DennyZhang has joined #openstack-dev | 09:32 | |
| *** neeti has quit IRC | 09:32 | |
| lifeless | I'm talking overhead on the wire | 09:34 |
| lifeless | we make lots of API responses between services | 09:34 |
| lifeless | like nova -> keystone | 09:34 |
| lifeless | extra fat in the response should have some value to the user | 09:34 |
| *** exed has joined #openstack-dev | 09:35 | |
| *** vkozhukalov has quit IRC | 09:35 | |
| *** neeti has joined #openstack-dev | 09:35 | |
| lifeless | yolanda: anyhow, - highly recommend a list discussion | 09:37 |
| yolanda | lifeless, i just sent the email now | 09:37 |
| lifeless | cool, thanks | 09:37 |
| *** Abhishek_ has joined #openstack-dev | 09:39 | |
| *** ffio has quit IRC | 09:39 | |
| *** noorul has quit IRC | 09:41 | |
| shardy | Does anyone from stable-maint know why Heat 2013.2.1 has been released without merging all the fixes queued for stable/havana? | 09:41 |
| *** nelsnelson has quit IRC | 09:41 | |
| *** nelsnelson has joined #openstack-dev | 09:42 | |
| *** andreaf has quit IRC | 09:43 | |
| *** renlt has quit IRC | 09:43 | |
| *** martyntaylor has joined #openstack-dev | 09:44 | |
| *** bingbu has quit IRC | 09:46 | |
| *** danpb has joined #openstack-dev | 09:46 | |
| *** faramir1 has quit IRC | 09:46 | |
| *** martyntaylor1 has joined #openstack-dev | 09:47 | |
| *** rohitk has joined #openstack-dev | 09:47 | |
| *** teran has joined #openstack-dev | 09:47 | |
| *** martyntaylor has quit IRC | 09:48 | |
| *** noorul has joined #openstack-dev | 09:50 | |
| *** vkozhukalov has joined #openstack-dev | 09:50 | |
| *** haypo has left #openstack-dev | 09:51 | |
| *** boris-42 has joined #openstack-dev | 09:52 | |
| *** teran has quit IRC | 09:52 | |
| *** nati_ueno has quit IRC | 09:53 | |
| *** DennyZhang has quit IRC | 09:54 | |
| *** nati_ueno has joined #openstack-dev | 09:55 | |
| *** SergeyLukjanov has quit IRC | 09:55 | |
| *** SergeyLukjanov has joined #openstack-dev | 09:57 | |
| *** teran has joined #openstack-dev | 09:58 | |
| *** JordanP has joined #openstack-dev | 09:59 | |
| *** max_lobur_afk is now known as max_lobur | 10:00 | |
| *** iartarisi has joined #openstack-dev | 10:00 | |
| *** doron_afk is now known as doron | 10:00 | |
| *** teran has quit IRC | 10:02 | |
| *** pixelb has joined #openstack-dev | 10:03 | |
| *** faramir1 has joined #openstack-dev | 10:04 | |
| *** diakunchikov__ has quit IRC | 10:06 | |
| *** aditirav has quit IRC | 10:07 | |
| *** aditirav has joined #openstack-dev | 10:09 | |
| *** sahid has joined #openstack-dev | 10:10 | |
| *** rushiagr has quit IRC | 10:13 | |
| *** xingchao has quit IRC | 10:13 | |
| *** comay has quit IRC | 10:13 | |
| *** comay has joined #openstack-dev | 10:14 | |
| *** rushiagr has joined #openstack-dev | 10:14 | |
| *** rcleere has quit IRC | 10:15 | |
| *** rcleere has joined #openstack-dev | 10:15 | |
| *** itooon has quit IRC | 10:17 | |
| *** itooon has joined #openstack-dev | 10:20 | |
| *** krast has quit IRC | 10:21 | |
| *** neeti has quit IRC | 10:22 | |
| *** nshaikh has joined #openstack-dev | 10:22 | |
| *** neeti has joined #openstack-dev | 10:22 | |
| *** rushiagr has quit IRC | 10:25 | |
| *** Shaan7 has quit IRC | 10:27 | |
| *** doron is now known as doron_afk | 10:28 | |
| *** teran has joined #openstack-dev | 10:29 | |
| *** paragan has quit IRC | 10:31 | |
| *** jcooley_ has joined #openstack-dev | 10:33 | |
| *** rushiagr has joined #openstack-dev | 10:33 | |
| *** jcooley_ has quit IRC | 10:38 | |
| *** rushiagr2 has joined #openstack-dev | 10:39 | |
| *** vkozhukalov has quit IRC | 10:40 | |
| *** rushiagr has quit IRC | 10:40 | |
| *** jp_at_hp has joined #openstack-dev | 10:41 | |
| *** tonyfy has quit IRC | 10:42 | |
| *** tkammer has quit IRC | 10:42 | |
| *** tkammer has joined #openstack-dev | 10:42 | |
| *** jcooley_ has joined #openstack-dev | 10:43 | |
| *** rods has joined #openstack-dev | 10:44 | |
| *** sushil_ has joined #openstack-dev | 10:44 | |
| *** mrunge has quit IRC | 10:47 | |
| *** zul has quit IRC | 10:47 | |
| *** jcooley_ has quit IRC | 10:48 | |
| *** jcooley_ has joined #openstack-dev | 10:48 | |
| *** xqueralt has joined #openstack-dev | 10:48 | |
| *** zul has joined #openstack-dev | 10:48 | |
| *** sushil_ has quit IRC | 10:49 | |
| *** martyntaylor1 has quit IRC | 10:50 | |
| *** athomas has quit IRC | 10:51 | |
| *** vkozhukalov has joined #openstack-dev | 10:52 | |
| *** jcooley_ has quit IRC | 10:52 | |
| *** sushil_ has joined #openstack-dev | 10:53 | |
| *** mrunge has joined #openstack-dev | 10:53 | |
| *** jcooley_ has joined #openstack-dev | 10:53 | |
| *** xingchao has joined #openstack-dev | 10:54 | |
| *** athomas has joined #openstack-dev | 10:55 | |
| *** jaimegil has quit IRC | 10:57 | |
| *** jcooley_ has quit IRC | 10:58 | |
| *** jcooley_ has joined #openstack-dev | 10:58 | |
| *** xingchao_ has joined #openstack-dev | 10:59 | |
| *** SergeyLukjanov has quit IRC | 11:01 | |
| *** xingchao has quit IRC | 11:03 | |
| *** jcooley_ has quit IRC | 11:03 | |
| *** itooon has quit IRC | 11:03 | |
| pixelb | jd__ re ceilometer havana wsme cap. Would this commit be compat with wsme-0.5b5 or does it need 0.5b6? https://github.com/openstack/ceilometer/commit/e4a1a4f | 11:03 |
| *** jcooley_ has joined #openstack-dev | 11:04 | |
| *** rushiagr2 has quit IRC | 11:05 | |
| *** rushiagr3 has joined #openstack-dev | 11:05 | |
| *** guohliu has quit IRC | 11:05 | |
| *** flaper87 is now known as flaper87|afk | 11:07 | |
| *** jcooley_ has quit IRC | 11:08 | |
| *** ruhe has joined #openstack-dev | 11:08 | |
| *** xingchao_ has quit IRC | 11:08 | |
| *** evgenyf has quit IRC | 11:09 | |
| *** markmc has joined #openstack-dev | 11:09 | |
| *** amuller_ has joined #openstack-dev | 11:10 | |
| *** nati_ueno has quit IRC | 11:10 | |
| *** nati_ueno has joined #openstack-dev | 11:11 | |
| *** ArminderS- has quit IRC | 11:11 | |
| *** DinaBelova has quit IRC | 11:11 | |
| *** irenab has quit IRC | 11:11 | |
| *** amuller has quit IRC | 11:11 | |
| *** teju_ has joined #openstack-dev | 11:12 | |
| *** akrivoka has joined #openstack-dev | 11:12 | |
| *** nati_uen_ has joined #openstack-dev | 11:12 | |
| *** rushiagr3 is now known as rushiagr | 11:14 | |
| *** pcm has joined #openstack-dev | 11:14 | |
| *** safchain has quit IRC | 11:14 | |
| *** jcooley_ has joined #openstack-dev | 11:14 | |
| *** nati_ueno has quit IRC | 11:15 | |
| *** reidrac has joined #openstack-dev | 11:16 | |
| *** pcm has quit IRC | 11:16 | |
| pixelb | jd__, I guess not since it removes some validation code | 11:16 |
| *** pcm has joined #openstack-dev | 11:16 | |
| *** evgenyf has joined #openstack-dev | 11:16 | |
| *** marun has joined #openstack-dev | 11:17 | |
| *** jimjiang has quit IRC | 11:17 | |
| *** jimjiang has joined #openstack-dev | 11:18 | |
| *** jcooley_ has quit IRC | 11:19 | |
| *** jcooley_ has joined #openstack-dev | 11:19 | |
| *** rongze has quit IRC | 11:19 | |
| *** teju_ has quit IRC | 11:20 | |
| *** jcooley_ has quit IRC | 11:23 | |
| *** jcooley_ has joined #openstack-dev | 11:24 | |
| *** jimjiang has quit IRC | 11:25 | |
| *** DinaBelova has joined #openstack-dev | 11:25 | |
| *** jimjiang has joined #openstack-dev | 11:26 | |
| *** SergeyLukjanov has joined #openstack-dev | 11:26 | |
| *** kushal has joined #openstack-dev | 11:27 | |
| *** salv-orlando has joined #openstack-dev | 11:28 | |
| *** jcooley_ has quit IRC | 11:29 | |
| *** jcooley_ has joined #openstack-dev | 11:30 | |
| *** jimjiang has quit IRC | 11:31 | |
| *** SergeyLukjanov has quit IRC | 11:34 | |
| *** jcooley_ has quit IRC | 11:34 | |
| *** jcooley_ has joined #openstack-dev | 11:35 | |
| *** rfolco has joined #openstack-dev | 11:36 | |
| *** unicell_ has quit IRC | 11:36 | |
| *** unicell has joined #openstack-dev | 11:36 | |
| *** unicell has quit IRC | 11:36 | |
| *** unicell has joined #openstack-dev | 11:36 | |
| *** SergeyLukjanov has joined #openstack-dev | 11:37 | |
| *** noorul has quit IRC | 11:37 | |
| *** xqueralt has quit IRC | 11:38 | |
| *** xqueralt has joined #openstack-dev | 11:38 | |
| *** jcooley_ has quit IRC | 11:39 | |
| *** jcooley_ has joined #openstack-dev | 11:40 | |
| *** zoresvit has joined #openstack-dev | 11:41 | |
| *** jimjiang has joined #openstack-dev | 11:42 | |
| *** ruhe has quit IRC | 11:42 | |
| *** ruhe has joined #openstack-dev | 11:43 | |
| *** rongze has joined #openstack-dev | 11:44 | |
| *** jcooley_ has quit IRC | 11:45 | |
| *** rongze_ has joined #openstack-dev | 11:45 | |
| *** ArxCruz has joined #openstack-dev | 11:45 | |
| *** jcooley_ has joined #openstack-dev | 11:46 | |
| *** salv-orlando has quit IRC | 11:47 | |
| *** jaimegil has joined #openstack-dev | 11:47 | |
| *** rods has quit IRC | 11:48 | |
| *** venum has quit IRC | 11:48 | |
| *** ruhe has quit IRC | 11:48 | |
| *** zaneb has quit IRC | 11:49 | |
| *** rongze has quit IRC | 11:49 | |
| *** Alexei_987 has joined #openstack-dev | 11:50 | |
| *** itooon has joined #openstack-dev | 11:50 | |
| *** HenryG has quit IRC | 11:50 | |
| *** jcooley_ has quit IRC | 11:50 | |
| *** jcooley_ has joined #openstack-dev | 11:51 | |
| *** jimjiang has quit IRC | 11:51 | |
| *** jimjiang has joined #openstack-dev | 11:53 | |
| *** chandankumar has quit IRC | 11:53 | |
| *** chandankumar has joined #openstack-dev | 11:53 | |
| *** kutija has joined #openstack-dev | 11:53 | |
| marun | uh, anybody else having issues with restacking due to nova-* services running as daemon and persisting past screen being killed? | 11:54 |
| *** jcooley_ has quit IRC | 11:55 | |
| *** jcooley_ has joined #openstack-dev | 11:56 | |
| *** jimjiang has quit IRC | 11:57 | |
| *** jimjiang has joined #openstack-dev | 11:58 | |
| *** drewlander has joined #openstack-dev | 11:58 | |
| *** mkollaro has joined #openstack-dev | 11:59 | |
| *** jcooley_ has quit IRC | 12:01 | |
| *** jcooley_ has joined #openstack-dev | 12:01 | |
| *** zaneb has joined #openstack-dev | 12:02 | |
| *** noorul has joined #openstack-dev | 12:02 | |
| *** CaptTofu has quit IRC | 12:02 | |
| *** CaptTofu has joined #openstack-dev | 12:03 | |
| *** jimjiang has quit IRC | 12:03 | |
| *** rongze_ has quit IRC | 12:03 | |
| *** jimjiang has joined #openstack-dev | 12:04 | |
| *** lucasagomes is now known as lucas-hungry | 12:05 | |
| *** teran_ has joined #openstack-dev | 12:05 | |
| *** jcooley_ has quit IRC | 12:05 | |
| *** shakayumi has joined #openstack-dev | 12:06 | |
| *** teran has quit IRC | 12:08 | |
| *** rushiagr2 has joined #openstack-dev | 12:08 | |
| *** rushiagr has quit IRC | 12:09 | |
| *** salv-orlando has joined #openstack-dev | 12:10 | |
| *** salv-orlando has quit IRC | 12:10 | |
| *** glenng has quit IRC | 12:11 | |
| *** jimjiang has quit IRC | 12:11 | |
| *** jcooley_ has joined #openstack-dev | 12:11 | |
| *** rushiagr3 has joined #openstack-dev | 12:11 | |
| *** jimjiang has joined #openstack-dev | 12:12 | |
| *** rushiagr2 has quit IRC | 12:13 | |
| *** corXi has quit IRC | 12:13 | |
| *** corXi has joined #openstack-dev | 12:13 | |
| *** CaptTofu has quit IRC | 12:14 | |
| *** CaptTofu has joined #openstack-dev | 12:14 | |
| *** jcooley_ has quit IRC | 12:15 | |
| *** dstanek has quit IRC | 12:16 | |
| *** jcooley_ has joined #openstack-dev | 12:16 | |
| *** akscram has quit IRC | 12:17 | |
| *** Abhishek_ has quit IRC | 12:17 | |
| *** hajay_ has joined #openstack-dev | 12:19 | |
| *** jcooley_ has quit IRC | 12:21 | |
| *** jcooley_ has joined #openstack-dev | 12:21 | |
| *** jaimegil has quit IRC | 12:22 | |
| *** martyntaylor has joined #openstack-dev | 12:22 | |
| *** dave_tucker has joined #openstack-dev | 12:24 | |
| *** jcooley_ has quit IRC | 12:25 | |
| *** corXi has quit IRC | 12:26 | |
| *** corXi has joined #openstack-dev | 12:26 | |
| *** jcooley_ has joined #openstack-dev | 12:26 | |
| *** jruzicka has joined #openstack-dev | 12:27 | |
| *** jaimegil has joined #openstack-dev | 12:27 | |
| *** FunnyLookinHat has joined #openstack-dev | 12:27 | |
| *** faramir1 has quit IRC | 12:28 | |
| *** nshaikh has left #openstack-dev | 12:28 | |
| *** dolphm has joined #openstack-dev | 12:29 | |
| *** mkollaro1 has joined #openstack-dev | 12:30 | |
| *** mkollaro has quit IRC | 12:30 | |
| *** jcooley_ has quit IRC | 12:30 | |
| *** jcooley_ has joined #openstack-dev | 12:31 | |
| *** devvesa has quit IRC | 12:32 | |
| *** marios_ has joined #openstack-dev | 12:33 | |
| *** marios_ has quit IRC | 12:33 | |
| *** yaguang has quit IRC | 12:33 | |
| *** paragan has joined #openstack-dev | 12:33 | |
| *** rongze has joined #openstack-dev | 12:34 | |
| *** yolanda has quit IRC | 12:35 | |
| *** jaimegil has quit IRC | 12:35 | |
| *** yolanda has joined #openstack-dev | 12:36 | |
| *** tdruiva has joined #openstack-dev | 12:36 | |
| *** jcooley_ has quit IRC | 12:36 | |
| *** jcooley_ has joined #openstack-dev | 12:37 | |
| *** jcooley_ has quit IRC | 12:37 | |
| *** tdruiva has quit IRC | 12:37 | |
| *** kutija has quit IRC | 12:37 | |
| *** jcooley_ has joined #openstack-dev | 12:38 | |
| *** tdruiva has joined #openstack-dev | 12:38 | |
| *** corXi has quit IRC | 12:38 | |
| *** e0ne has quit IRC | 12:38 | |
| *** corXi has joined #openstack-dev | 12:40 | |
| *** morazi has joined #openstack-dev | 12:40 | |
| *** rongze has quit IRC | 12:42 | |
| *** jcooley_ has quit IRC | 12:42 | |
| *** tdruiva has quit IRC | 12:42 | |
| *** jcooley_ has joined #openstack-dev | 12:43 | |
| *** jcooley_ has quit IRC | 12:43 | |
| *** abramley has quit IRC | 12:45 | |
| *** jcooley_ has joined #openstack-dev | 12:49 | |
| *** gszasz has quit IRC | 12:50 | |
| *** nosnos has quit IRC | 12:51 | |
| *** nurla has joined #openstack-dev | 12:51 | |
| *** tdruiva has joined #openstack-dev | 12:52 | |
| *** athomas has quit IRC | 12:53 | |
| *** gszasz has joined #openstack-dev | 12:53 | |
| *** jcooley_ has quit IRC | 12:54 | |
| *** andreaf has joined #openstack-dev | 12:54 | |
| *** jcooley_ has joined #openstack-dev | 12:54 | |
| *** exed has quit IRC | 12:54 | |
| *** irenab has joined #openstack-dev | 12:54 | |
| *** aditirav has quit IRC | 12:55 | |
| *** SergeyLukjanov has quit IRC | 12:56 | |
| *** mkollaro has joined #openstack-dev | 12:57 | |
| *** kutija has joined #openstack-dev | 12:58 | |
| *** SergeyLukjanov has joined #openstack-dev | 12:58 | |
| *** jimjiang has quit IRC | 12:59 | |
| *** jcooley_ has quit IRC | 12:59 | |
| *** jcooley_ has joined #openstack-dev | 13:00 | |
| *** exed has joined #openstack-dev | 13:00 | |
| *** jimjiang has joined #openstack-dev | 13:00 | |
| *** athomas has joined #openstack-dev | 13:00 | |
| *** mkollaro1 has quit IRC | 13:00 | |
| ttx | markmcclain: around? | 13:01 |
| markmcclain | ttx: hi | 13:01 |
| ttx | still doing 1:1s here, no time to set up my logged office channel | 13:01 |
| markmcclain | ok | 13:01 |
| *** mkollaro has quit IRC | 13:02 | |
| ttx | https://launchpad.net/neutron/+milestone/icehouse-2 | 13:02 |
| *** gszasz has quit IRC | 13:02 | |
| *** afazekas has quit IRC | 13:02 | |
| ttx | only one to triage: https://blueprints.launchpad.net/neutron/+spec/floating-ip-extra-route | 13:02 |
| markmcclain | yeah… still on the fence on that one | 13:02 |
| ttx | looks like you beat the tide | 13:02 |
| markmcclain | yeah | 13:03 |
| ttx | Good progress too overall | 13:03 |
| *** sthaha has joined #openstack-dev | 13:03 | |
| *** sthaha has quit IRC | 13:03 | |
| *** sthaha has joined #openstack-dev | 13:03 | |
| markmcclain | yeah we have a healthy mix in review now which is good | 13:03 |
| ttx | markmcclain: a few dependency issues that might be worth exploring: | 13:04 |
| *** xga has quit IRC | 13:04 | |
| *** xga_ has joined #openstack-dev | 13:04 | |
| ttx | dnsmasq-mode-keyword (Medium, scollins, icehouse-2) depends on ipv6-feature-parity (Undefined, markmcclain, icehouse-3) ? | 13:04 |
| *** jcooley_ has quit IRC | 13:04 | |
| markmcclain | oops.. I'll fix that | 13:04 |
| *** jimjiang has quit IRC | 13:04 | |
| *** gargya has joined #openstack-dev | 13:04 | |
| ttx | same for vif-attribute-for-hairpinning (Medium, scollins, icehouse-2) | 13:04 |
| *** jcooley_ has joined #openstack-dev | 13:04 | |
| ttx | same for ipv6-provider-nets-slaac (Low, comcast-openstack, icehouse-2) | 13:04 |
| ttx | all depending on ipv6-feature-parity | 13:04 |
| markmcclain | ah all of the ipv6 ones | 13:04 |
| ttx | was ipv6-feature-parity recently moved to i3 ? | 13:05 |
| *** jimjiang has joined #openstack-dev | 13:05 | |
| markmcclain | not intentionally | 13:06 |
| *** yassine has quit IRC | 13:06 | |
| markmcclain | may have accidentally selected wrong milestone in the dropdown | 13:06 |
| *** rushiagr3 has quit IRC | 13:06 | |
| *** nati_uen_ has quit IRC | 13:06 | |
| *** yassine has joined #openstack-dev | 13:06 | |
| ttx | markmcclain: yeah, sounds like something you should at least ttry to get in i2... otherwise could slip (again) | 13:06 |
| *** mkollaro has joined #openstack-dev | 13:07 | |
| markmcclain | yeah… I'll also reparent a few to make the dep a cleaner | 13:07 |
| ttx | markmcclain: once that's fixed it all looks good. If you get bored during the holidays you can start triaging the i3 blueprints | 13:08 |
| markmcclain | a gift to myself :) | 13:08 |
| ttx | markmcclain: is there anything you'd like to discuss at the meeting today ? | 13:08 |
| markmcclain | no | 13:08 |
| markmcclain | no 1:1 next Tuesday right? | 13:09 |
| ttx | We'll skip the next two meetings, I think | 13:09 |
| *** jcooley_ has quit IRC | 13:09 | |
| ttx | 24 and 31... | 13:09 |
| markmcclain | ok | 13:09 |
| *** e0ne has joined #openstack-dev | 13:09 | |
| ttx | so we'll skip the 1:1s as well | 13:09 |
| ttx | markmcclain: see you later! | 13:09 |
| *** xingchao has joined #openstack-dev | 13:09 | |
| markmcclain | talk to you this evening | 13:09 |
| *** jcooley_ has joined #openstack-dev | 13:10 | |
| *** shakayumi has quit IRC | 13:10 | |
| *** jimjiang has quit IRC | 13:11 | |
| *** shakayumi has joined #openstack-dev | 13:11 | |
| *** _dbalog has quit IRC | 13:12 | |
| *** jimjiang has joined #openstack-dev | 13:12 | |
| *** flaper87|afk is now known as flaper87 | 13:13 | |
| *** e0ne has quit IRC | 13:14 | |
| *** jcooley_ has quit IRC | 13:14 | |
| *** jcooley_ has joined #openstack-dev | 13:14 | |
| ttx | dhellmann: around? | 13:15 |
| dhellmann | hi, ttx! | 13:15 |
| *** haomaiwang has quit IRC | 13:15 | |
| ttx | dhellmann: good morning! | 13:15 |
| ttx | https://launchpad.net/oslo/+milestone/icehouse-2 | 13:15 |
| dhellmann | ttx: good afternoon! | 13:15 |
| *** haomaiwang has joined #openstack-dev | 13:15 | |
| ttx | https://blueprints.launchpad.net/oslo/+spec/middleware-request-id needs a priority (or be booted) | 13:15 |
| *** shakayumi has quit IRC | 13:15 | |
| *** ruhe has joined #openstack-dev | 13:16 | |
| ttx | otherwise looks good | 13:16 |
| dhellmann | yeah, I was just looking at that one -- I need to track down the owner | 13:16 |
| ttx | https://launchpad.net/oslo.messaging/+milestone/icehouse-2 looks good | 13:16 |
| dhellmann | markmc has done a good job keeping that organized | 13:16 |
| ttx | A bit too many "not started" I think, with the holidays coming up | 13:17 |
| ttx | (on the "oslo" side) | 13:17 |
| *** yamahata has joined #openstack-dev | 13:17 | |
| dhellmann | yeah, a few of those are mine, and I expect to have some time to start the high priority one this week | 13:17 |
| ttx | ok | 13:17 |
| dhellmann | the others may slide, but I'll give them until the first of the year in case people are planning to work on them during the holidays | 13:17 |
| *** jimjiang has quit IRC | 13:17 | |
| ttx | improve-config-discovery-for-docs (Medium, doug-hellmann, icehouse-2) depends on remove-import-side-effects (Undefined, doug-hellmann, No milestone) ? | 13:17 |
| *** rongze has joined #openstack-dev | 13:18 | |
| dhellmann | hmm, those are related but I don't think the dependency is strictly true | 13:18 |
| dhellmann | maybe it's backwards | 13:18 |
| *** dolphm has quit IRC | 13:18 | |
| dhellmann | the config discovery stuff can (and should) be done before the other work, so I guess I set that relationship up wrong | 13:18 |
| ttx | dhellmann: ok, I'll fix it | 13:18 |
| dhellmann | sorry about that, I didn't notice I'd done that | 13:19 |
| *** jimjiang has joined #openstack-dev | 13:19 | |
| *** mrunge has quit IRC | 13:19 | |
| *** jcooley_ has quit IRC | 13:19 | |
| ttx | otherwise all looks good | 13:19 |
| ttx | dhellmann: anything you want to discuss at the cross-project meeting later ? | 13:19 |
| *** jcooley_ has joined #openstack-dev | 13:19 | |
| dhellmann | no, I don't think so | 13:20 |
| ttx | We'll skip the next two 1:1s and cross-project meeting fwiw | 13:20 |
| ttx | (on the 24 and the 31) | 13:20 |
| dhellmann | ok, makes sense | 13:20 |
| *** urulama has quit IRC | 13:20 | |
| ttx | dhellmann: ok, then I'll see you at the meeting(s) later | 13:21 |
| dhellmann | ok, thanks, see you then | 13:21 |
| ttx | thx! | 13:21 |
| *** thomasem has joined #openstack-dev | 13:21 | |
| *** haomaiwa_ has joined #openstack-dev | 13:22 | |
| *** afazekas has joined #openstack-dev | 13:22 | |
| *** dolphm has joined #openstack-dev | 13:23 | |
| *** ilyashakhat has joined #openstack-dev | 13:23 | |
| *** haomaiwang has quit IRC | 13:24 | |
| *** jimjiang has quit IRC | 13:24 | |
| *** haomaiwa_ has quit IRC | 13:24 | |
| *** jcooley_ has quit IRC | 13:24 | |
| *** ilyashakhat_ has quit IRC | 13:24 | |
| *** e0ne has joined #openstack-dev | 13:25 | |
| *** tdruiva has quit IRC | 13:25 | |
| *** tdruiva has joined #openstack-dev | 13:26 | |
| *** jcooley_ has joined #openstack-dev | 13:26 | |
| *** dolphm has quit IRC | 13:27 | |
| *** sandywalsh_ has quit IRC | 13:29 | |
| *** ruhe is now known as ruhe_ | 13:29 | |
| *** jcooley_ has quit IRC | 13:30 | |
| *** jimjiang has joined #openstack-dev | 13:30 | |
| *** ruhe_ is now known as ruhe | 13:31 | |
| *** haomaiwang has joined #openstack-dev | 13:31 | |
| *** doron_afk is now known as doron | 13:31 | |
| *** jcooley_ has joined #openstack-dev | 13:32 | |
| *** rods has joined #openstack-dev | 13:32 | |
| *** jdob has joined #openstack-dev | 13:35 | |
| *** jimjiang has quit IRC | 13:35 | |
| *** jimjiang has joined #openstack-dev | 13:36 | |
| *** jcooley_ has quit IRC | 13:36 | |
| *** boris-42 has quit IRC | 13:37 | |
| *** ig has quit IRC | 13:37 | |
| *** jcooley_ has joined #openstack-dev | 13:37 | |
| *** Shaan7 has joined #openstack-dev | 13:37 | |
| *** Shaan7 has quit IRC | 13:37 | |
| *** Shaan7 has joined #openstack-dev | 13:37 | |
| *** martyntaylor has quit IRC | 13:38 | |
| *** ruhe is now known as ruhe_ | 13:38 | |
| *** buzztroll has joined #openstack-dev | 13:38 | |
| *** ig has joined #openstack-dev | 13:39 | |
| *** Nikolay_1t has quit IRC | 13:39 | |
| *** jobewan_ has joined #openstack-dev | 13:39 | |
| *** jobewan_ has quit IRC | 13:39 | |
| *** rongze_ has joined #openstack-dev | 13:39 | |
| *** jobewan_ has joined #openstack-dev | 13:40 | |
| *** ig has quit IRC | 13:40 | |
| *** jobewan_ has quit IRC | 13:40 | |
| *** ruhe_ is now known as ruhe | 13:41 | |
| *** ig has joined #openstack-dev | 13:41 | |
| *** jimjiang has quit IRC | 13:41 | |
| *** sandywalsh_ has joined #openstack-dev | 13:42 | |
| *** jimjiang has joined #openstack-dev | 13:42 | |
| *** AMike has quit IRC | 13:42 | |
| *** jcooley_ has quit IRC | 13:42 | |
| *** AMike has joined #openstack-dev | 13:43 | |
| *** AMike has quit IRC | 13:43 | |
| *** AMike has joined #openstack-dev | 13:43 | |
| *** jcooley_ has joined #openstack-dev | 13:43 | |
| *** rongze has quit IRC | 13:43 | |
| *** jobewan has joined #openstack-dev | 13:43 | |
| *** tosate has joined #openstack-dev | 13:44 | |
| *** gargya_ has joined #openstack-dev | 13:44 | |
| *** gargya__ has joined #openstack-dev | 13:45 | |
| *** gargya has quit IRC | 13:45 | |
| *** gargya__ is now known as gargya | 13:46 | |
| ttx | jgriffith: around? | 13:47 |
| *** rongze_ has quit IRC | 13:48 | |
| *** jcooley_ has quit IRC | 13:48 | |
| *** jcooley_ has joined #openstack-dev | 13:48 | |
| *** rongze has joined #openstack-dev | 13:48 | |
| *** tosate has quit IRC | 13:48 | |
| *** irenab has quit IRC | 13:48 | |
| *** gargya_ has quit IRC | 13:48 | |
| *** buzztroll has quit IRC | 13:48 | |
| *** rongze_ has joined #openstack-dev | 13:49 | |
| *** jimjiang has quit IRC | 13:49 | |
| *** dolphm has joined #openstack-dev | 13:49 | |
| *** dsirrine has joined #openstack-dev | 13:49 | |
| *** dvarga has joined #openstack-dev | 13:50 | |
| *** martyntaylor has joined #openstack-dev | 13:50 | |
| *** mkollaro1 has joined #openstack-dev | 13:51 | |
| *** mkollaro has quit IRC | 13:51 | |
| *** dolphm has quit IRC | 13:51 | |
| *** dolphm_ has joined #openstack-dev | 13:51 | |
| *** afazekas has quit IRC | 13:51 | |
| *** jcooley_ has quit IRC | 13:52 | |
| *** rongze has quit IRC | 13:53 | |
| *** shakayumi has joined #openstack-dev | 13:54 | |
| *** dprince has joined #openstack-dev | 13:54 | |
| *** jcooley_ has joined #openstack-dev | 13:54 | |
| *** buzztroll has joined #openstack-dev | 13:54 | |
| *** jcooley_ has quit IRC | 13:54 | |
| *** asselin has quit IRC | 13:55 | |
| *** asselin has joined #openstack-dev | 13:55 | |
| *** itooon has quit IRC | 13:55 | |
| *** martyntaylor has quit IRC | 13:56 | |
| *** tris has quit IRC | 13:56 | |
| *** itooon has joined #openstack-dev | 13:56 | |
| *** jcooley_ has joined #openstack-dev | 13:56 | |
| *** xga has joined #openstack-dev | 13:56 | |
| *** xga_ has quit IRC | 13:56 | |
| *** rfolco has quit IRC | 13:56 | |
| *** pfreund has quit IRC | 13:56 | |
| *** lsmola_ has quit IRC | 13:56 | |
| *** jimjiang has joined #openstack-dev | 13:57 | |
| *** corXi has quit IRC | 13:58 | |
| *** gargya_ has joined #openstack-dev | 13:58 | |
| *** corXi has joined #openstack-dev | 13:58 | |
| *** dolphm_ has quit IRC | 13:58 | |
| *** lxsli has joined #openstack-dev | 13:58 | |
| *** gargya has quit IRC | 13:59 | |
| *** cpallares has joined #openstack-dev | 13:59 | |
| *** gargya_ is now known as gargya | 13:59 | |
| *** HenryG has joined #openstack-dev | 14:00 | |
| *** corXi has quit IRC | 14:00 | |
| *** glenng has joined #openstack-dev | 14:00 | |
| *** corXi has joined #openstack-dev | 14:01 | |
| *** jcooley_ has quit IRC | 14:01 | |
| *** jimjiang has quit IRC | 14:01 | |
| *** corXi has joined #openstack-dev | 14:01 | |
| *** jcooley_ has joined #openstack-dev | 14:01 | |
| *** tris has joined #openstack-dev | 14:01 | |
| *** jimjiang has joined #openstack-dev | 14:01 | |
| *** gszasz has joined #openstack-dev | 14:02 | |
| *** gargya_ has joined #openstack-dev | 14:02 | |
| *** martyntaylor has joined #openstack-dev | 14:03 | |
| *** yaguang has joined #openstack-dev | 14:03 | |
| *** dolphm has joined #openstack-dev | 14:03 | |
| *** sthaha has quit IRC | 14:04 | |
| *** jdob has quit IRC | 14:04 | |
| *** jdob_ has joined #openstack-dev | 14:04 | |
| *** hajay__ has joined #openstack-dev | 14:05 | |
| *** gargya has quit IRC | 14:05 | |
| *** gargya_ is now known as gargya | 14:06 | |
| *** jcooley_ has quit IRC | 14:06 | |
| *** aeperezt has joined #openstack-dev | 14:06 | |
| *** jcooley_ has joined #openstack-dev | 14:07 | |
| *** heyongli has joined #openstack-dev | 14:07 | |
| *** e0ne_ has joined #openstack-dev | 14:07 | |
| *** dolphm has quit IRC | 14:08 | |
| *** rfolco has joined #openstack-dev | 14:08 | |
| *** hajay_ has quit IRC | 14:08 | |
| *** MaxV has joined #openstack-dev | 14:09 | |
| *** hajay__ has quit IRC | 14:09 | |
| *** lsmola_ has joined #openstack-dev | 14:10 | |
| *** pfreund has joined #openstack-dev | 14:11 | |
| *** e0ne has quit IRC | 14:11 | |
| *** alop has joined #openstack-dev | 14:11 | |
| *** jcooley_ has quit IRC | 14:12 | |
| *** jcooley_ has joined #openstack-dev | 14:13 | |
| *** akscram has joined #openstack-dev | 14:13 | |
| *** clayb has joined #openstack-dev | 14:14 | |
| *** eharney has joined #openstack-dev | 14:15 | |
| *** jimjiang has quit IRC | 14:16 | |
| *** jcooley_ has quit IRC | 14:17 | |
| *** jimjiang has joined #openstack-dev | 14:17 | |
| *** abramley has joined #openstack-dev | 14:17 | |
| *** stannie has joined #openstack-dev | 14:18 | |
| *** jcooley_ has joined #openstack-dev | 14:18 | |
| *** kbringard has joined #openstack-dev | 14:18 | |
| *** jcooley_ has quit IRC | 14:18 | |
| *** joesavak has joined #openstack-dev | 14:18 | |
| *** jdob_ has quit IRC | 14:18 | |
| *** jdob has joined #openstack-dev | 14:18 | |
| *** nermina has joined #openstack-dev | 14:19 | |
| *** jcooley_ has joined #openstack-dev | 14:20 | |
| *** jimjiang has quit IRC | 14:22 | |
| *** danielbruno has joined #openstack-dev | 14:22 | |
| *** afazekas has joined #openstack-dev | 14:23 | |
| *** doron is now known as doron_afk | 14:23 | |
| *** jimjiang has joined #openstack-dev | 14:23 | |
| *** jcooley_ has quit IRC | 14:24 | |
| *** tris has quit IRC | 14:25 | |
| *** dolphm has joined #openstack-dev | 14:25 | |
| *** jcooley_ has joined #openstack-dev | 14:25 | |
| *** tris has joined #openstack-dev | 14:25 | |
| *** radez_g0n3 is now known as radez | 14:26 | |
| *** KurtMartin has joined #openstack-dev | 14:26 | |
| *** jimjiang has quit IRC | 14:27 | |
| *** yamahata__ has quit IRC | 14:27 | |
| *** abramley has quit IRC | 14:28 | |
| *** mindpixel has quit IRC | 14:28 | |
| *** yamahata__ has joined #openstack-dev | 14:28 | |
| *** jcooley_ has quit IRC | 14:29 | |
| *** jcooley_ has joined #openstack-dev | 14:30 | |
| *** mindpixel has joined #openstack-dev | 14:30 | |
| jgriffith | ttx: hola | 14:30 |
| *** andreaf has quit IRC | 14:31 | |
| *** ruhe has quit IRC | 14:31 | |
| *** shakayumi has quit IRC | 14:31 | |
| *** yamahata has quit IRC | 14:32 | |
| ttx | jgriffith: we can do your 1:1 now if you want (originally scheduled at 13:45 UTC) | 14:33 |
| jgriffith | ttx: yes please | 14:33 |
| *** jimjiang has joined #openstack-dev | 14:33 | |
| *** markmcclain has quit IRC | 14:33 | |
| *** yamahata has joined #openstack-dev | 14:34 | |
| ttx | https://launchpad.net/cinder/+milestone/icehouse-2 | 14:34 |
| jgriffith | ttx: sorry... thought we werwe 14:45 | 14:34 |
| ttx | looks good, one left to triage @ https://blueprints.launchpad.net/cinder/+spec/cinder-backup-recover-api | 14:34 |
| *** jcooley_ has quit IRC | 14:34 | |
| *** mriedem has joined #openstack-dev | 14:34 | |
| jgriffith | ttx: on it | 14:35 |
| ttx | jgriffith: there is also a dependency issue: | 14:35 |
| jgriffith | ttx: which one? | 14:35 |
| ttx | copy-volume-to-image-task-flow (Low, abhijeet-malawade, icehouse-2) depends on create-volume-flow (Undefined, anastasia-karpinska, No milestone) | 14:35 |
| jgriffith | aye | 14:36 |
| jgriffith | taskflow, smaskflow | 14:36 |
| *** Shaan7 has quit IRC | 14:36 | |
| jgriffith | ttx: I still need to sync with those folks | 14:36 |
| jgriffith | ttx: I'm a bit unclear of their plans here TBH | 14:36 |
| *** Shaan7 has joined #openstack-dev | 14:37 | |
| * jgriffith is making a note to sort these at tomorrows meeting | 14:37 | |
| ttx | jgriffith: cool, that's about it. Good progress overall | 14:37 |
| ttx | we'll skip the next 2 1:1s and cross-project meetings | 14:37 |
| jgriffith | Yeah, I'm happy with how things are going the milestone | 14:37 |
| jgriffith | indeed... | 14:37 |
| jgriffith | Happy Holidays! | 14:37 |
| ttx | is there anything you want to see discussed at the meeting today ? | 14:37 |
| jgriffith | Nope | 14:38 |
| jgriffith | saving my troulbe for the new year ;) | 14:38 |
| jgriffith | trouble | 14:38 |
| ttx | jgriffith: ok then, see you there | 14:38 |
| *** abramley has joined #openstack-dev | 14:39 | |
| *** davidhadas has joined #openstack-dev | 14:39 | |
| jgriffith | ttx: thanks! | 14:39 |
| *** jcooley_ has joined #openstack-dev | 14:39 | |
| *** rushiagr3 has joined #openstack-dev | 14:41 | |
| *** jayg|g0n3 is now known as jayg | 14:42 | |
| *** xingchao has quit IRC | 14:43 | |
| *** jcooley_ has quit IRC | 14:44 | |
| *** ruhe has joined #openstack-dev | 14:44 | |
| *** rohitk has quit IRC | 14:44 | |
| *** FunnyLookinHat has quit IRC | 14:45 | |
| *** jcooley_ has joined #openstack-dev | 14:45 | |
| *** glenng has quit IRC | 14:45 | |
| *** glenng has joined #openstack-dev | 14:45 | |
| *** n0ano has joined #openstack-dev | 14:45 | |
| *** heyongli has quit IRC | 14:45 | |
| *** jasondotstar has joined #openstack-dev | 14:46 | |
| *** yolanda has quit IRC | 14:48 | |
| *** burt has joined #openstack-dev | 14:48 | |
| *** mtreinish has quit IRC | 14:48 | |
| *** alpha_ori has quit IRC | 14:48 | |
| jgriffith | sdague: ic you assigned yourself: https://bugs.launchpad.net/cinder/+bug/1261731 | 14:49 |
| uvirtbot | Launchpad bug 1261731 in cinder "Unable to deactivate LVM snapshot" [Undecided,In progress] | 14:49 |
| *** pixelb has quit IRC | 14:49 | |
| *** yolanda has joined #openstack-dev | 14:49 | |
| jgriffith | sdague: assuming just change: https://github.com/openstack/cinder/blob/master/cinder/brick/local_dev/lvm.py#L577 to info | 14:49 |
| jgriffith | sdague: I'll leave it to you, if you don't have time lemme know | 14:49 |
| *** jcooley_ has quit IRC | 14:49 | |
| *** topol has joined #openstack-dev | 14:49 | |
| sdague | jgriffith: there is actually a patch out there | 14:50 |
| *** dperaza has joined #openstack-dev | 14:50 | |
| sdague | jgriffith: https://review.openstack.org/#/c/62622/ | 14:50 |
| sdague | I changed them to debug | 14:50 |
| *** pixelb has joined #openstack-dev | 14:50 | |
| sdague | because if it all just works, then the end user doesn't really care | 14:50 |
| *** redbo has quit IRC | 14:50 | |
| sdague | only devs that want to see that it tried a recovery | 14:50 |
| *** jcooley_ has joined #openstack-dev | 14:51 | |
| *** redbo has joined #openstack-dev | 14:51 | |
| *** heyongli has joined #openstack-dev | 14:51 | |
| *** mtreinish has joined #openstack-dev | 14:51 | |
| *** alpha_ori has joined #openstack-dev | 14:51 | |
| *** ayoung has joined #openstack-dev | 14:51 | |
| *** nkinder has quit IRC | 14:52 | |
| *** aron28 has joined #openstack-dev | 14:52 | |
| *** aron28 is now known as syed_armani | 14:53 | |
| *** syed_armani has left #openstack-dev | 14:53 | |
| *** syed_armani has joined #openstack-dev | 14:54 | |
| *** buzztroll has quit IRC | 14:54 | |
| *** mfer has joined #openstack-dev | 14:54 | |
| *** jcooley_ has quit IRC | 14:55 | |
| *** mindpixel has quit IRC | 14:55 | |
| *** ffio has joined #openstack-dev | 14:55 | |
| *** jcooley_ has joined #openstack-dev | 14:55 | |
| *** jecarey has joined #openstack-dev | 14:57 | |
| *** gargya has quit IRC | 14:57 | |
| *** david-lyle_ has joined #openstack-dev | 14:57 | |
| *** xingchao has joined #openstack-dev | 14:58 | |
| *** _dbalog has joined #openstack-dev | 14:58 | |
| *** chandankumar has quit IRC | 14:59 | |
| *** achampion has quit IRC | 14:59 | |
| *** e0ne_ is now known as e0ne | 15:00 | |
| syed_armani | tristanC: ping | 15:00 |
| *** achampion has joined #openstack-dev | 15:00 | |
| *** xingchao has quit IRC | 15:00 | |
| *** jcooley_ has quit IRC | 15:00 | |
| ttx | dolphm: around? | 15:00 |
| *** markmcclain has joined #openstack-dev | 15:01 | |
| *** mriedem has quit IRC | 15:01 | |
| tristanC | syed_armani: pong! | 15:01 |
| *** Ryan_Lane has joined #openstack-dev | 15:01 | |
| dolphm | ttx: got rescheduled to do the project status update webinar thing right now :) | 15:01 |
| syed_armani | tristanC: I am wondering if you are working on this bug --> https://bugs.launchpad.net/horizon/+bug/1202965 | 15:01 |
| *** tkammer has quit IRC | 15:01 | |
| uvirtbot | Launchpad bug 1202965 in nova "The name VCPUs (total) of Hypervisors is confusing" [Undecided,In progress] | 15:01 |
| dolphm | ttx: i'll poke you when i'm done? | 15:01 |
| ttx | dolphm: yay | 15:01 |
| ttx | dolphm: yes, we'll see when we can talk | 15:01 |
| tristanC | syed_armani: oups, I forgot to de/assigned myself. I tried to but my change is more complex than excepted :( | 15:02 |
| ttx | russellb: interested in stealing the slot ? | 15:02 |
| tristanC | syed_armani: so no | 15:02 |
| *** mriedem has joined #openstack-dev | 15:02 | |
| *** jcooley_ has joined #openstack-dev | 15:02 | |
| syed_armani | tristanC: :( , thx!! | 15:02 |
| *** buzztroll has joined #openstack-dev | 15:03 | |
| russellb | ttx: hey, sure we can go ahead and start | 15:03 |
| ttx | https://launchpad.net/nova/+milestone/icehouse-2 | 15:03 |
| russellb | i have a deadline set for this thursday | 15:03 |
| *** matiu has quit IRC | 15:03 | |
| russellb | for getting blueprints approved for icehouse-2 | 15:03 |
| ttx | russellb: ok | 15:03 |
| russellb | the rest will be deferred | 15:03 |
| *** vijendar has joined #openstack-dev | 15:03 | |
| *** aditirav has joined #openstack-dev | 15:04 | |
| russellb | lots of critical bugs unassigned, too ... | 15:04 |
| russellb | all gate bugs though | 15:04 |
| russellb | and people don't really like to assign themselves, even if they are pitching in debugging | 15:04 |
| ttx | more work needed on rpc-major-version-updates-icehouse ? | 15:04 |
| russellb | yes, that should be set back to good progress | 15:04 |
| russellb | will fix | 15:04 |
| *** xga has quit IRC | 15:05 | |
| russellb | it was all done, but then i abandoned some of it and have to redo it | 15:05 |
| *** xingchao has joined #openstack-dev | 15:05 | |
| *** davidhadas has quit IRC | 15:05 | |
| russellb | changed | 15:05 |
| ttx | russellb: ok, I think we'll see clearer after Thursday | 15:05 |
| russellb | hope so | 15:05 |
| ttx | maybe I'll ping you again then, so that we can have it clean and tidy for the holidays | 15:05 |
| russellb | sure that's fine | 15:06 |
| russellb | probably Friday | 15:06 |
| ttx | because next 2 weeks we'll skip 1:1s and meetings | 15:06 |
| russellb | since EOD is deadline | 15:06 |
| russellb | if you're around friday | 15:06 |
| ttx | sure. Ping me if i forgot about it | 15:06 |
| russellb | heh will try | 15:06 |
| russellb | i can put it on my calendar | 15:06 |
| *** mindpixel has joined #openstack-dev | 15:07 | |
| ttx | russellb: little point in discussing progress until that's done. | 15:07 |
| ttx | russellb: anything you want to discuss at the meeting today ? | 15:07 |
| *** matiu has joined #openstack-dev | 15:07 | |
| *** doron_afk is now known as doron | 15:08 | |
| *** neeti has quit IRC | 15:08 | |
| *** davidhadas has joined #openstack-dev | 15:08 | |
| *** xga has joined #openstack-dev | 15:08 | |
| russellb | i don't think so | 15:08 |
| *** david-lyle_ has quit IRC | 15:08 | |
| ttx | russellb: ok then, see you there | 15:08 |
| russellb | thanks | 15:08 |
| *** david-lyle_ has joined #openstack-dev | 15:08 | |
| notmyname | russellb: have you seen http://not.mn/gate_status.html ? | 15:09 |
| ttx | david-lyle: we can talk now if you're around | 15:10 |
| david-lyle_ | sure, here | 15:10 |
| russellb | notmyname: just saw it a bit ago this morning | 15:10 |
| russellb | notmyname: presumably you follow the "state of the gate" reports on the list? | 15:10 |
| ttx | david-lyle: https://launchpad.net/horizon/+milestone/icehouse-2 | 15:10 |
| ttx | david-lyle: status looks good | 15:11 |
| notmyname | russellb: yes | 15:11 |
| ttx | david-lyle: progress is good | 15:11 |
| *** rushiagr3 is now known as rushiagr | 15:11 | |
| *** alex_xu has quit IRC | 15:11 | |
| ttx | david-lyle: we might have trouble discussing the dependency issue with ceilometer, as jd__ will miss the meeting today | 15:12 |
| david-lyle_ | ttx: several bps are close to landing | 15:12 |
| *** stevemar has joined #openstack-dev | 15:12 | |
| jd__ | ttx: nah I might be there | 15:12 |
| ttx | jd__: oh, cool | 15:12 |
| ttx | FTR: horizon/ceilometer-api-enhancements <- ceilometer/statistics-order-by-and-limit-for-grouped-query | 15:12 |
| *** safchain has joined #openstack-dev | 15:13 | |
| ttx | jd__: or we can discuss it now if you prefer | 15:13 |
| jd__ | ttx: I've 10 minutes if that's enough :) | 15:13 |
| *** KurtMartin has quit IRC | 15:13 | |
| *** KurtMartin has joined #openstack-dev | 15:14 | |
| ttx | we can try | 15:14 |
| *** KurtMartin has quit IRC | 15:14 | |
| ttx | david-lyle: what do you need from ceilometer to complete that one ? | 15:14 |
| david-lyle_ | ttx: I've asked lsmola to split the Horizon bp into two parts, one that depends on the ceilometer bp, and one that is the remainder | 15:14 |
| *** otherwiseguy has joined #openstack-dev | 15:15 | |
| ttx | and do the dependent one once the feature hits in ceilometer ? | 15:15 |
| david-lyle_ | yes, that way it's not completely blocked | 15:15 |
| ttx | (i.e. target the dependent one to "next") | 15:15 |
| david-lyle_ | yes | 15:15 |
| ttx | david-lyle: that works | 15:15 |
| *** alunduil has quit IRC | 15:16 | |
| *** kevinconway has joined #openstack-dev | 15:16 | |
| *** hajay_ has joined #openstack-dev | 15:16 | |
| ttx | jd__: that statistics-order-by-and-limit-for-grouped-query won't get done in icehouse, right ? | 15:16 |
| david-lyle_ | we'll see if we can make that split happen today | 15:16 |
| ttx | david-lyle: that would be great, thx | 15:16 |
| *** ngoracke has joined #openstack-dev | 15:16 | |
| *** gszasz has quit IRC | 15:16 | |
| ttx | david-lyle: anything you'd like to see discussed at the meeting today ? | 15:16 |
| ttx | david-lyle: we'll skip the next 2 weeks 1:1s and meetings, fwiw | 15:16 |
| *** FunnyLookinHat has joined #openstack-dev | 15:17 | |
| jd__ | ttx: no assignee, so I won't count on it | 15:17 |
| ttx | jd__: ok | 15:17 |
| *** gszasz has joined #openstack-dev | 15:17 | |
| david-lyle_ | ttx: I think we're ok, just need to keep making progress | 15:17 |
| *** ngoracke1 has joined #openstack-dev | 15:17 | |
| ttx | david-lyle: thx! see you at meeting later | 15:17 |
| david-lyle_ | ttx: thanks | 15:18 |
| david-lyle_ | dolphm: what's the proper way to change the author on an existing gerrit review? | 15:18 |
| david-lyle_ | --amend --author doesn't seem to work | 15:19 |
| *** hajay_ has quit IRC | 15:20 | |
| *** dave_tucker has quit IRC | 15:20 | |
| *** urulama has joined #openstack-dev | 15:21 | |
| *** lucas-hungry is now known as lucasagomes | 15:21 | |
| *** ngoracke has quit IRC | 15:21 | |
| *** nelsnelson has quit IRC | 15:21 | |
| *** nelsnelson has joined #openstack-dev | 15:21 | |
| *** pfreund has quit IRC | 15:22 | |
| *** heyongli has quit IRC | 15:23 | |
| *** neelashah has joined #openstack-dev | 15:25 | |
| *** urulama has quit IRC | 15:25 | |
| *** AlanClark has joined #openstack-dev | 15:25 | |
| *** yamahata has quit IRC | 15:26 | |
| *** spzala has joined #openstack-dev | 15:27 | |
| *** DinaBelova has quit IRC | 15:30 | |
| *** ffio has quit IRC | 15:31 | |
| *** drewlander has quit IRC | 15:32 | |
| *** amuller_ has quit IRC | 15:33 | |
| *** rnirmal has joined #openstack-dev | 15:34 | |
| *** ayoung has quit IRC | 15:34 | |
| *** mindpixeler has joined #openstack-dev | 15:39 | |
| *** tanisdl has joined #openstack-dev | 15:39 | |
| *** sarob has joined #openstack-dev | 15:40 | |
| *** rollietikes has joined #openstack-dev | 15:40 | |
| *** sarob has quit IRC | 15:40 | |
| stevemar | ayoung: what should we do about bug 1261622 | 15:40 |
| uvirtbot | Launchpad bug 1261622 in keystone "change text or behaviour of the admin token in keystone.conf" [High,Confirmed] https://launchpad.net/bugs/1261622 | 15:40 |
| *** mikeoutland has joined #openstack-dev | 15:40 | |
| *** sarob has joined #openstack-dev | 15:40 | |
| *** luisg has quit IRC | 15:40 | |
| *** reed has joined #openstack-dev | 15:40 | |
| *** mindpixel has quit IRC | 15:42 | |
| *** DinaBelova has joined #openstack-dev | 15:42 | |
| *** sarob has quit IRC | 15:43 | |
| *** tkammer has joined #openstack-dev | 15:43 | |
| *** CaptTofu has quit IRC | 15:44 | |
| *** jergerber has joined #openstack-dev | 15:45 | |
| *** carl_baldwin has joined #openstack-dev | 15:45 | |
| *** rraja has quit IRC | 15:45 | |
| *** sarob has joined #openstack-dev | 15:45 | |
| dolphm | david-lyle_: that's exactly what should work | 15:46 |
| dolphm | david-lyle_: actually gerrit probably won't reflect the new authorship for the review level, but the latest patchset should reflect it | 15:47 |
| *** jcoufal has quit IRC | 15:47 | |
| *** sarob has quit IRC | 15:48 | |
| dolphm | ttx: o/ free | 15:48 |
| *** sarob_ has joined #openstack-dev | 15:48 | |
| david-lyle_ | dolphm: ok, issue has been launchpad expiring the review while being worked on, so just changing the author might not work | 15:48 |
| *** SergeyLukjanov has quit IRC | 15:48 | |
| dolphm | david-lyle_: expiring the review or bug? | 15:48 |
| *** DinaBelova has quit IRC | 15:48 | |
| dolphm | david-lyle_: or gerrit expiring the review after a downvote? | 15:48 |
| david-lyle_ | dolphm: that latter | 15:49 |
| david-lyle_ | the | 15:49 |
| ttx | dolphm: let's try to do it quickly now ? | 15:49 |
| dolphm | david-lyle_: i'm familiar with that problem :P maybe -ci can bump the expiry up from a week? | 15:49 |
| *** bswartz has quit IRC | 15:49 | |
| dolphm | ttx: sure, if you have time | 15:49 |
| david-lyle_ | dolphm: thanks | 15:49 |
| ttx | 10min | 15:49 |
| *** corXi has quit IRC | 15:49 | |
| *** luisg has joined #openstack-dev | 15:49 | |
| ttx | dolphm: https://launchpad.net/keystone/+milestone/icehouse-2 | 15:50 |
| dolphm | i still owe you federation blueprint dependency unwinding btw | 15:50 |
| ttx | dolphm: still two to prioritize in there | 15:50 |
| ttx | https://blueprints.launchpad.net/keystone/+spec/service-scoped-tokens | 15:50 |
| ttx | https://blueprints.launchpad.net/keystone/+spec/name-spaced-roles | 15:50 |
| *** mlavalle has joined #openstack-dev | 15:50 | |
| *** alex_klimov has quit IRC | 15:50 | |
| *** julienvey_ has quit IRC | 15:50 | |
| *** vkozhukalov has quit IRC | 15:51 | |
| *** corXi has joined #openstack-dev | 15:51 | |
| *** aeperezt has quit IRC | 15:51 | |
| ttx | dolphm: there is another dep issue with: | 15:52 |
| ttx | saml-id (Medium, marek-denis, icehouse-2) depending on auth-domain-architecture (Undefined, No assignee, No milestone) | 15:52 |
| *** doron is now known as doron_afk | 15:52 | |
| dolphm | that's the federation-related bp unwinding i owe you :) | 15:52 |
| ttx | oh, that too. ok :) | 15:53 |
| *** sarob_ has quit IRC | 15:53 | |
| ttx | otherwise progress looks good | 15:53 |
| *** salv-orlando has joined #openstack-dev | 15:53 | |
| thomasem | jaypipes, Hey, would you have any time today to chat a bit about some schema concerns that were brought to my attention for our event/trait tables in Ceilometer? :) | 15:53 |
| ttx | dolphm: anything you want to discuss at meeting later ? | 15:53 |
| dolphm | i don't believe so | 15:53 |
| ttx | dolphm: we'll skip next two weeks (1:1s and meeting) | 15:53 |
| dolphm | ah cool | 15:53 |
| jaypipes | thomasem: absolutely. in about an hour work? | 15:53 |
| thomasem | jaypipes, Wanted to get your thoughts since we're in the process of load testing the drivers. That'd work great! | 15:54 |
| ttx | dolphm: ok, so those two are the only things you need to fix (untriaged stuff and federation unwind) | 15:54 |
| ttx | dolphm: see you at meeting later | 15:54 |
| dolphm | ttx: /salute | 15:54 |
| ttx | thx! | 15:54 |
| *** alexpilotti has joined #openstack-dev | 15:54 | |
| jaypipes | thomasem: sounds good. ttyl. | 15:54 |
| *** mriedem has quit IRC | 15:55 | |
| *** nkinder has joined #openstack-dev | 15:55 | |
| dolphm | dhellmann: global requirements patch you might be interested in https://review.openstack.org/#/c/61198/ | 15:55 |
| *** aditirav_ has joined #openstack-dev | 15:56 | |
| *** jecarey has quit IRC | 15:56 | |
| *** dolphm is now known as dolphm_afk | 15:57 | |
| *** johnthetubaguy1 has joined #openstack-dev | 15:57 | |
| *** viktors has joined #openstack-dev | 15:58 | |
| *** aditirav has quit IRC | 15:58 | |
| *** johnthetubaguy has quit IRC | 15:58 | |
| *** boris-42 has joined #openstack-dev | 15:59 | |
| mfer | can anyone direct me to a standard on api versioning, features, and how trunk is always usable? | 15:59 |
| *** xingchao has quit IRC | 15:59 | |
| mfer | i'm trying to understand how new features affect apis, how the versions change, and how to write clients against this | 16:00 |
| *** aditirav_ has quit IRC | 16:00 | |
| *** Ruetobas has quit IRC | 16:01 | |
| *** pmathews has joined #openstack-dev | 16:01 | |
| *** alunduil has joined #openstack-dev | 16:01 | |
| *** bdpayne has joined #openstack-dev | 16:02 | |
| *** radez is now known as radez_g0n3 | 16:02 | |
| *** bnemec_ is now known as bnemec | 16:02 | |
| sdague | annegentle_: you about. | 16:02 |
| *** eharney has quit IRC | 16:03 | |
| *** aditirav has joined #openstack-dev | 16:03 | |
| *** Ruetobas has joined #openstack-dev | 16:03 | |
| *** eharney has joined #openstack-dev | 16:03 | |
| *** noorul has quit IRC | 16:03 | |
| *** drewlander has joined #openstack-dev | 16:03 | |
| *** aeperezt has joined #openstack-dev | 16:03 | |
| annegentle_ | sdague: here! | 16:03 |
| sdague | so I continue to be confused by the commit message request | 16:03 |
| *** reidrac has quit IRC | 16:04 | |
| sdague | openstack/governace: publish docs to docs.openstack.org | 16:04 |
| annegentle_ | sdague: it looks wide-scoped to me | 16:04 |
| annegentle_ | sdague: I want it narrowly scoped in the git history | 16:04 |
| sdague | seems pretty clear that it would be governance docs | 16:04 |
| annegentle_ | sdague: even with the metadata | 16:04 |
| annegentle_ | sdague: it gave me pause and might do the same to others | 16:04 |
| topol | sdague, is there a URL that describes how many builds OpenStack does a day. I remember hearing you quote a value. I cant remember what it was | 16:04 |
| sdague | so this infrastructure would allow you to publish anything out of the repo | 16:05 |
| sdague | though nothing is currently published, because we need to restructure (or add symlinks) to make that work | 16:05 |
| annegentle_ | sdague: yes but we wouldn't convert all docs | 16:05 |
| sdague | annegentle_: maybe, maybe not. This change doesn't really get affected either way | 16:05 |
| sdague | topol: http://dague.net/2013/10/27/openstack-ci-by-the-numbers/ was a point in time | 16:07 |
| *** Ruetobas has quit IRC | 16:07 | |
| *** drewlander has quit IRC | 16:07 | |
| *** teju_ has joined #openstack-dev | 16:08 | |
| topol | sdague, is there a good number I can quote for number of builds a day? | 16:10 |
| topol | sdague, wasnt it like 800? | 16:10 |
| sdague | the most recent numbers I've got are the ones above, and they are really scoped to the week level | 16:11 |
| sdague | you could go mine elastic search, that's what I did | 16:12 |
| *** Guest53164 is now known as antonym | 16:12 | |
| *** mikeoutland has quit IRC | 16:13 | |
| *** Ruetobas has joined #openstack-dev | 16:13 | |
| *** rnirmal has quit IRC | 16:13 | |
| *** hcc has joined #openstack-dev | 16:14 | |
| *** martyntaylor has quit IRC | 16:14 | |
| *** rnirmal has joined #openstack-dev | 16:14 | |
| *** DennyZhang has joined #openstack-dev | 16:14 | |
| *** HenryG has quit IRC | 16:14 | |
| *** jistr has quit IRC | 16:14 | |
| *** HenryG has joined #openstack-dev | 16:14 | |
| *** markwash has joined #openstack-dev | 16:16 | |
| annegentle_ | sdague: in the land of governance I think it's best to scope definitively even in meta messaging like commit messages. Not digging my heels in, but being particular about scope and wording for 15 years from now. Our future selves will appreciate the detail. | 16:16 |
| sdague | annegentle_: but this change doesn't actually specify in any way what content will get web published | 16:17 |
| *** garyk has quit IRC | 16:17 | |
| sdague | it's actually generic infrastructure for a docs jobs | 16:18 |
| *** hcc has quit IRC | 16:18 | |
| sdague | content tbd | 16:18 |
| annegentle_ | sdague: then I want to define what content is intended in the commit message. Unless your intentions are for all docs? | 16:18 |
| sdague | annegentle_: my intention is for content decisions to be made on their own merrits in the future | 16:18 |
| annegentle_ | sdague: I'd also really prefer we investigate build jobs to the wiki, but there's not really a way to discuss that other than -2 on that review... | 16:19 |
| annegentle_ | sdague: which seems overbearing | 16:19 |
| sdague | so I think that is a worthy discussion that should come onto the agenda if you want to | 16:20 |
| annegentle_ | sdague: I can bring it up at the TC | 16:20 |
| *** omachace has left #openstack-dev | 16:20 | |
| annegentle_ | sdague: when we review the patch (which is on the agenda) | 16:20 |
| sdague | I'm honestly pretty negative on publish to the wiki | 16:20 |
| annegentle_ | sdague: that's cool, we'll discuss | 16:20 |
| sdague | because I think we actually have a really solid workflow of docs, with review, with anyone able to contribute to go to static trees | 16:21 |
| *** paragan has quit IRC | 16:21 | |
| sdague | sure | 16:21 |
| *** buzztroll has quit IRC | 16:21 | |
| *** sstent has joined #openstack-dev | 16:22 | |
| sstent | Anyone seen this error in conductor before? Expected exception during message handling (Could not find binary nova-compute on host compute-1) | 16:22 |
| *** mdomsch has quit IRC | 16:22 | |
| *** buzztroll has joined #openstack-dev | 16:24 | |
| *** DennyZhang has quit IRC | 16:24 | |
| *** jecarey has joined #openstack-dev | 16:24 | |
| *** DennyZhang has joined #openstack-dev | 16:24 | |
| *** tan has quit IRC | 16:24 | |
| *** ifarkas has quit IRC | 16:25 | |
| *** markmcclain has quit IRC | 16:26 | |
| *** xqueralt has quit IRC | 16:26 | |
| *** akrivoka has quit IRC | 16:27 | |
| *** martyntaylor has joined #openstack-dev | 16:27 | |
| *** iartarisi has quit IRC | 16:28 | |
| *** kenperkins has joined #openstack-dev | 16:28 | |
| *** corXi has quit IRC | 16:28 | |
| *** jecarey_ has joined #openstack-dev | 16:28 | |
| *** jcooley_ has quit IRC | 16:30 | |
| *** teju_ has quit IRC | 16:31 | |
| *** xga has quit IRC | 16:31 | |
| *** jecarey has quit IRC | 16:31 | |
| *** DennyZhang has quit IRC | 16:33 | |
| *** doron_afk is now known as doron | 16:33 | |
| *** DennyZhang has joined #openstack-dev | 16:33 | |
| *** MaxV has quit IRC | 16:34 | |
| *** pberis has quit IRC | 16:34 | |
| *** dolphm_afk has quit IRC | 16:35 | |
| *** CaptTofu has joined #openstack-dev | 16:35 | |
| *** xga has joined #openstack-dev | 16:35 | |
| *** herndon has joined #openstack-dev | 16:35 | |
| *** ayoung has joined #openstack-dev | 16:35 | |
| *** rcleere has quit IRC | 16:37 | |
| *** MaxV has joined #openstack-dev | 16:37 | |
| *** Mandell has quit IRC | 16:37 | |
| *** CaptTofu has quit IRC | 16:37 | |
| *** CaptTofu has joined #openstack-dev | 16:37 | |
| *** CaptTofu has quit IRC | 16:38 | |
| *** CaptTofu has joined #openstack-dev | 16:38 | |
| *** adalbas has quit IRC | 16:41 | |
| sstent | Anyone seen this error in conductor before? Expected exception during message handling (Could not find binary nova-compute on host compute-1) | 16:42 |
| *** martyntaylor has quit IRC | 16:43 | |
| *** mrodden1 has quit IRC | 16:44 | |
| *** doron is now known as doron_afk | 16:44 | |
| *** BLZbubba_ has quit IRC | 16:44 | |
| *** arozumenko has quit IRC | 16:44 | |
| *** arozumenko has joined #openstack-dev | 16:44 | |
| markwash | ttx: good evening | 16:45 |
| *** hemna has joined #openstack-dev | 16:45 | |
| *** BLZbubba has joined #openstack-dev | 16:46 | |
| *** mindpixeler has quit IRC | 16:47 | |
| ttx | markwash: hi! | 16:48 |
| *** doron_afk has quit IRC | 16:48 | |
| ttx | https://launchpad.net/glance/+milestone/icehouse-2 | 16:48 |
| ttx | looks good | 16:49 |
| markwash | cool | 16:49 |
| *** mindpixel has joined #openstack-dev | 16:49 | |
| markwash | I guess now is our weekly opportunity to highlight mordred | 16:49 |
| markwash | but he's not even here! | 16:49 |
| ttx | hiding under that mtaylor guy I think | 16:50 |
| markwash | ah | 16:50 |
| ttx | anything you want to discuss at meeting today ? | 16:50 |
| markwash | nothing that comes to mind | 16:50 |
| ttx | we'll skip next 2 weeks of 1:1s & meetings fwiw | 16:51 |
| markwash | ttx: okay sounds fine | 16:51 |
| ttx | ok see you at meeting today ! | 16:51 |
| markwash | laters | 16:52 |
| *** tkammer has quit IRC | 16:53 | |
| *** yaguang has quit IRC | 16:53 | |
| *** rongze_ has quit IRC | 16:53 | |
| *** mindpixel has quit IRC | 16:53 | |
| *** rongze has joined #openstack-dev | 16:53 | |
| *** adalbas has joined #openstack-dev | 16:54 | |
| jaypipes | thomasem: arighty. ready when you are. | 16:54 |
| *** SergeyLukjanov has joined #openstack-dev | 16:54 | |
| *** mkollaro1 has quit IRC | 16:55 | |
| ttx | hub_cap: around? | 16:56 |
| *** vijendar has quit IRC | 16:56 | |
| *** vijendar has joined #openstack-dev | 16:56 | |
| hub_cap | ttx aye | 16:57 |
| hub_cap | lets burn thru this :) | 16:57 |
| ttx | https://launchpad.net/trove/+milestone/icehouse-2 | 16:57 |
| ttx | looks good | 16:57 |
| ttx | good progress too | 16:57 |
| hub_cap | yes some things are trickling in | 16:57 |
| hub_cap | ive added a new i3 bp in the last ~30 min | 16:58 |
| hub_cap | ive nuked all the essential bps down to high | 16:58 |
| ttx | you might want to set assignees for all blueprints on https://blueprints.launchpad.net/trove/icehouse | 16:58 |
| hub_cap | roger | 16:58 |
| ttx | nothing else, all looks good for the next two weeks | 16:59 |
| *** rcleere has joined #openstack-dev | 16:59 | |
| *** afazekas has quit IRC | 16:59 | |
| *** aditirav has quit IRC | 16:59 | |
| ttx | (we won't have 1:1s or cross-project meetings for the next 2 weeks) | 16:59 |
| ttx | hub_cap: anything you want to see discussed at the cross-project meeting later ? | 16:59 |
| *** drewlander has joined #openstack-dev | 16:59 | |
| hub_cap | so we wont have it 2x? | 16:59 |
| *** angdraug has joined #openstack-dev | 16:59 | |
| hub_cap | or skip one week? | 16:59 |
| ttx | hub_cap: yes. You fly solo | 16:59 |
| ttx | 2x | 16:59 |
| hub_cap | hehe cool | 17:00 |
| hub_cap | hopefully my autopilot will work, like in airplane | 17:00 |
| ttx | no interest in tracking status early in milestones anyway | 17:00 |
| hub_cap | sue | 17:00 |
| hub_cap | sure | 17:00 |
| ttx | (once the roadmap is set) | 17:00 |
| hub_cap | ok well awesome! im feeling good, nothing to really mention at todays meeting | 17:00 |
| *** simo is now known as simo_out | 17:01 | |
| hub_cap | i just hate bps ;) | 17:01 |
| ttx | hub_cap: any other question on milestoning, while I'm here ? | 17:01 |
| *** pberis has joined #openstack-dev | 17:01 | |
| *** haomaiwang has quit IRC | 17:02 | |
| *** pberis has quit IRC | 17:02 | |
| *** mrodden has joined #openstack-dev | 17:02 | |
| *** haomaiwang has joined #openstack-dev | 17:02 | |
| *** evgenyf has quit IRC | 17:03 | |
| *** stevemar has quit IRC | 17:03 | |
| ttx | hub_cap: ok, see you at meeting later | 17:03 |
| *** romcheg has quit IRC | 17:03 | |
| hub_cap | cu ttx | 17:04 |
| hub_cap | thx | 17:04 |
| *** xjiujiu has joined #openstack-dev | 17:04 | |
| *** devoid has joined #openstack-dev | 17:05 | |
| *** devoid has left #openstack-dev | 17:05 | |
| *** garyk has joined #openstack-dev | 17:05 | |
| *** gordc has joined #openstack-dev | 17:05 | |
| *** kenperkins has quit IRC | 17:05 | |
| *** ayoung has quit IRC | 17:06 | |
| *** pdevine has joined #openstack-dev | 17:06 | |
| *** noorul has joined #openstack-dev | 17:06 | |
| *** DennyZhang has quit IRC | 17:07 | |
| *** Abhishek_ has joined #openstack-dev | 17:07 | |
| *** haomaiwang has quit IRC | 17:07 | |
| *** davidhadas has quit IRC | 17:09 | |
| *** reed has quit IRC | 17:09 | |
| *** kbringard has quit IRC | 17:09 | |
| *** teran_ has quit IRC | 17:10 | |
| *** pablosan has joined #openstack-dev | 17:11 | |
| *** sushil_ has quit IRC | 17:13 | |
| shardy | giulivo: Hi | 17:13 |
| *** flaper87 is now known as flaper87|afk | 17:13 | |
| giulivo | shardy, I know you'd want to kill me now | 17:13 |
| *** markmc has quit IRC | 17:13 | |
| giulivo | I suspected that was going to happen, not only with you | 17:13 |
| shardy | giulivo: I'm just trying to understand what makes tempest special, in that you don't have to follow the same blueprint process as every other project? | 17:13 |
| giulivo | that probably *I* don't know the process | 17:14 |
| shardy | giulivo: Not a big deal, but we want to track getting the various heat tests into tempest, and BPs seem the best way to do that | 17:14 |
| shardy | giulivo: I can understand having etherpads for team task coordination, but IMHO they don't replace BPs | 17:15 |
| *** pablosan has quit IRC | 17:15 | |
| shardy | how will you track (accurately) what you deliver over the release cycle? | 17:15 |
| *** kbringard has joined #openstack-dev | 17:15 | |
| *** bswartz has joined #openstack-dev | 17:15 | |
| giulivo | I understant , this was actually a quite big issue with tempest as the general opinion is that blueprints aren't appropriate for new tests | 17:15 |
| *** saju_m has joined #openstack-dev | 17:16 | |
| giulivo | I think the general idea is that we should at least try to use only _a single_ blueprint per component when tracking new tests | 17:16 |
| *** e0ne has quit IRC | 17:16 | |
| giulivo | blueprints in tempest also have to deal with improvements to the framework itself | 17:16 |
| shardy | giulivo: Hmm. OK, that is, uh, very different to the process I'm familiar with | 17:17 |
| giulivo | but that won't allow to track new tests per release | 17:17 |
| shardy | giulivo: Can you link the discussion where this was decided please? | 17:17 |
| shardy | giulivo: No, you won't really be able to track anything, as you can't tag a commit with a line in an etherpad | 17:18 |
| *** sushils has joined #openstack-dev | 17:18 | |
| *** pablosan has joined #openstack-dev | 17:18 | |
| giulivo | shardy, so for scenarios, we have a single blueprint (across components) and that is targeted to a release | 17:18 |
| *** ayoung has joined #openstack-dev | 17:18 | |
| *** Shaan7 has quit IRC | 17:18 | |
| *** Shaan7 has joined #openstack-dev | 17:18 | |
| giulivo | for the api tests, tracking per component get things better organized but doesn't allow for good tracking | 17:19 |
| *** max_lobur is now known as max_lobur_afk | 17:19 | |
| *** stevemar has joined #openstack-dev | 17:19 | |
| shardy | giulivo: So if you land 200 new scenario tests, accross all openstack projects, that's tracked via *one* BP? | 17:19 |
| giulivo | shardy, yes that is | 17:19 |
| *** BLZbubba has quit IRC | 17:20 | |
| *** martyntaylor has joined #openstack-dev | 17:20 | |
| shardy | giulivo: OK, well let it be noted, I think you're making a mistake, but at least we now know what is expected | 17:20 |
| *** gyee has joined #openstack-dev | 17:20 | |
| shardy | giulivo: Is there an openstack-dev thread communicating this new process? | 17:21 |
| * shardy maybe missed it | 17:21 | |
| giulivo | shardy, yes I'll post a few links | 17:21 |
| giulivo | but honestly, for the purpose of tracking new tests, a single blueprint (maybe per component), per release, is quite good | 17:21 |
| *** ruhe has quit IRC | 17:21 | |
| *** dave_tucker has joined #openstack-dev | 17:21 | |
| giulivo | I don't see why spreading them around in more and more blueprints would help anyway | 17:21 |
| *** ffio has joined #openstack-dev | 17:21 | |
| *** urulama has joined #openstack-dev | 17:22 | |
| *** sushils has quit IRC | 17:22 | |
| *** ruhe has joined #openstack-dev | 17:22 | |
| giulivo | shardy, so this was introduced here: http://lists.openstack.org/pipermail/openstack-dev/2013-December/021922.html | 17:22 |
| *** viktors has left #openstack-dev | 17:23 | |
| *** sushils has joined #openstack-dev | 17:24 | |
| shardy | giulivo: Blueprints allow sharing tasks, and tracking those tasks (automatically through commit tags) | 17:24 |
| shardy | Neither of those things can be done accurately via an etherpad | 17:24 |
| *** nermina has quit IRC | 17:24 | |
| giulivo | I hear you on that, but I still don't think multiple blueprints would help either | 17:25 |
| *** kbringard1 has joined #openstack-dev | 17:25 | |
| *** saju_m has quit IRC | 17:25 | |
| *** glucas has joined #openstack-dev | 17:25 | |
| giulivo | for scenarios (which easily are cross-component) a single blueprint per release seems feasible | 17:25 |
| *** utlemming has quit IRC | 17:25 | |
| giulivo | for the single components, what do you think about a single bp per release, per components? | 17:26 |
| *** urulama has quit IRC | 17:26 | |
| *** kbringard has quit IRC | 17:27 | |
| *** ayoung has quit IRC | 17:27 | |
| *** mikeoutland has joined #openstack-dev | 17:27 | |
| giulivo | that is mainly the direction I tried to take now with heat, where I considered this https://blueprints.launchpad.net/tempest/+spec/tempest-heat-integration the "master plan" for the icehouse release | 17:27 |
| *** jswarren has joined #openstack-dev | 17:27 | |
| *** ayoung has joined #openstack-dev | 17:28 | |
| giulivo | shardy, I see as we speak more bp for heat are getting in | 17:29 |
| *** kbringard1 has quit IRC | 17:29 | |
| *** xjiujiu has quit IRC | 17:29 | |
| shardy | giulivo: We have been raising BP's, which were going to be linked as dependencies to the heat-integration umbrella BP | 17:30 |
| *** kbringard has joined #openstack-dev | 17:31 | |
| *** jruzicka has quit IRC | 17:31 | |
| shardy | then we can divide the various tasks up, allocate them to different people, and track the progress via launchpad | 17:31 |
| *** xjiujiu has joined #openstack-dev | 17:31 | |
| *** kenperkins has joined #openstack-dev | 17:31 | |
| *** exed has quit IRC | 17:31 | |
| shardy | giulivo: Or we can juggle multiple bit-rotted etherpads and have everyone get in an uncoordinated mess | 17:31 |
| *** kenperkins has quit IRC | 17:31 | |
| *** xarses has quit IRC | 17:32 | |
| shardy | giulivo: If this is the direction tempest has agreed, fair enough, but the granularity you propose to track things at makes no sense to me | 17:32 |
| *** andreaf has joined #openstack-dev | 17:32 | |
| shardy | one BP should be something you can allocate to one person, so you have a point of contact for status, and you avoid duplicated effort | 17:33 |
| giulivo | so I think we definitely don't want to have a blueprint per test, despite that providing greatest granularity | 17:33 |
| *** JordanP has quit IRC | 17:33 | |
| shardy | giulivo: So you would propose we just track stuff in an etherpad, and divide the allocation of the various tests up that way? | 17:34 |
| *** gothicmindfood has joined #openstack-dev | 17:34 | |
| *** s2r2 has quit IRC | 17:35 | |
| shardy | I understand the overhead involved with the BP/bug process is hard to manage, but IMO it does basically work, so short-circuiting it like this seems weird and counter-productive | 17:35 |
| *** exed has joined #openstack-dev | 17:36 | |
| giulivo | as you noted, there has been an attempt to use bugs, but then we had duplicates | 17:36 |
| *** martyntaylor has quit IRC | 17:37 | |
| *** Shaan7 has quit IRC | 17:37 | |
| *** ruhe has quit IRC | 17:37 | |
| giulivo | TODO items in the blueprint seem an intermediate measure; having blueprints which document tests that should be added aren't really useful, there is a lot of tests that could be added | 17:38 |
| *** s2r2 has joined #openstack-dev | 17:38 | |
| shardy | giulivo: If every project stopped doing issue tracking because they had duplicate bugs, there would be chaos ;) | 17:39 |
| shardy | giulivo: raising a bp can also be used as a statement of intent, by a developer, to do some work | 17:39 |
| *** rongze has quit IRC | 17:39 | |
| giulivo | shardy, indeed, that is the good way to use bp, I agree | 17:40 |
| *** rongze has joined #openstack-dev | 17:40 | |
| giulivo | so bugs were instead just leftovers with comments about missing tests, frequently duplicated by "super" blueprints | 17:40 |
| shardy | giulivo: I agree that randomly raising blueprints for wishlist items you have no intention of doing isn't helpful, but at least on heat most of our BPs aren't like that | 17:40 |
| shardy | giulivo: So in this case, most of the heat tempest BPs are a statement of intent, where we have decided to have a push and get some better Heat testing into tempest | 17:41 |
| *** d0ugal has quit IRC | 17:41 | |
| *** zoresvit has quit IRC | 17:41 | |
| *** MaxV has quit IRC | 17:41 | |
| *** mancdaz is now known as mancdaz_away | 17:42 | |
| shardy | giulivo: It would be valuable to get early feedback on the variouys tests (ie if they will be accepted), but if you feel that is too much overhead, we'll track them ourselves and just post the patches | 17:42 |
| *** exed has quit IRC | 17:42 | |
| *** dstanek has joined #openstack-dev | 17:42 | |
| *** mancdaz_away is now known as mancdaz | 17:42 | |
| shardy | That is the other benefit of BPs, details of what will be done, and discussion | 17:42 |
| giulivo | shardy, well I think this will turn useful if people contributing tests can use it | 17:42 |
| *** zoresvit has joined #openstack-dev | 17:43 | |
| shardy | giulivo: I am describing the process every openstack project uses, AFAIK | 17:43 |
| *** arnaud has joined #openstack-dev | 17:43 | |
| *** arnaud__ has joined #openstack-dev | 17:43 | |
| *** Underbyte has joined #openstack-dev | 17:43 | |
| *** jcooley_ has joined #openstack-dev | 17:43 | |
| giulivo | shardy, I got that but is that the case for "new tests"? we use that for new features in the framework or to enable support for new components | 17:43 |
| *** hemna has quit IRC | 17:44 | |
| *** vkozhukalov has joined #openstack-dev | 17:44 | |
| shardy | giulivo: I don't see why it can't be both - most of the code in the tempest tree is tests right? | 17:44 |
| *** utlemming has joined #openstack-dev | 17:44 | |
| *** hemna has joined #openstack-dev | 17:44 | |
| *** atiwari has joined #openstack-dev | 17:45 | |
| giulivo | shardy, sure it is so again, I found the intermediate solution of a single blueprint to fit well but, let me discuss that in the meeting so maybe we can come up with a stronger decision | 17:45 |
| *** ygbo has quit IRC | 17:46 | |
| *** markwash has quit IRC | 17:46 | |
| *** JordanRinke has quit IRC | 17:46 | |
| *** Drankis has joined #openstack-dev | 17:47 | |
| *** Drankis has quit IRC | 17:47 | |
| *** Drankis has joined #openstack-dev | 17:47 | |
| *** drankis_ has joined #openstack-dev | 17:47 | |
| shardy | giulivo: Ok, cool thanks | 17:47 |
| *** SumitNaiksatam has quit IRC | 17:48 | |
| shardy | giulivo: FWIW, my suggestion is to use BPs for new test development, but not to triage/approve those without an assignee | 17:48 |
| *** reed has joined #openstack-dev | 17:49 | |
| *** zoresvit has quit IRC | 17:49 | |
| *** sarob has joined #openstack-dev | 17:49 | |
| giulivo | shardy, agreed so the discussion would basically be about merging it in a single bp or not, correct? | 17:50 |
| *** sarob has quit IRC | 17:51 | |
| shardy | giulivo: Yes, I think it should be valid to allow folks to raise BPs for individual tests, then link them as dependencies to the "master plan" for a particular project | 17:51 |
| *** ndipanov has quit IRC | 17:51 | |
| *** jcooley_ has quit IRC | 17:51 | |
| *** abramley has quit IRC | 17:51 | |
| *** sarob has joined #openstack-dev | 17:51 | |
| shardy | giulivo: Not that everyone should use that process, of one giant BP works for them, but I think it's a valid and familiar workflow for many of us | 17:51 |
| *** mancdaz is now known as mancdaz_away | 17:51 | |
| shardy | s/of/if | 17:51 |
| *** jcooley_ has joined #openstack-dev | 17:51 | |
| *** eglynn has quit IRC | 17:52 | |
| *** danpb has quit IRC | 17:52 | |
| *** radix_ has joined #openstack-dev | 17:53 | |
| giulivo | thanks for pinging and taking some time to discuss it | 17:53 |
| *** JordanRinke has joined #openstack-dev | 17:53 | |
| *** rongze has quit IRC | 17:53 | |
| *** rongze has joined #openstack-dev | 17:54 | |
| *** sahid has quit IRC | 17:54 | |
| *** d0ugal has joined #openstack-dev | 17:54 | |
| *** d0ugal has quit IRC | 17:54 | |
| *** d0ugal has joined #openstack-dev | 17:54 | |
| *** gothicmindfood has quit IRC | 17:54 | |
| *** d0ugal has quit IRC | 17:55 | |
| *** nermina has joined #openstack-dev | 17:55 | |
| *** markmcclain has joined #openstack-dev | 17:55 | |
| *** xjiujiu has quit IRC | 17:55 | |
| *** sarob has quit IRC | 17:55 | |
| *** tanisdl_ has joined #openstack-dev | 17:56 | |
| *** jcooley_ has quit IRC | 17:56 | |
| *** tanisdl has quit IRC | 17:56 | |
| *** tanisdl_ is now known as tanisdl | 17:56 | |
| *** nati_ueno has joined #openstack-dev | 17:56 | |
| *** derekh has quit IRC | 17:58 | |
| *** syed_armani has quit IRC | 17:58 | |
| *** safchain has quit IRC | 17:58 | |
| morganfainberg | dstanek, looking at your comments, after meeting might discuss more-real-time than via gerrit | 17:59 |
| *** dolphm_afk has joined #openstack-dev | 17:59 | |
| dstanek | morganfainberg: sounds good | 18:00 |
| *** dolphm_afk is now known as dolphm | 18:00 | |
| *** rpodolyaka1 has joined #openstack-dev | 18:00 | |
| *** hajay_ has joined #openstack-dev | 18:00 | |
| *** xarses has joined #openstack-dev | 18:00 | |
| *** jecarey_ has quit IRC | 18:01 | |
| *** nati_uen_ has joined #openstack-dev | 18:02 | |
| *** jamielennox|away is now known as jamielennox | 18:03 | |
| *** amcrn has joined #openstack-dev | 18:03 | |
| *** nati_uen_ has quit IRC | 18:04 | |
| *** nati_uen_ has joined #openstack-dev | 18:04 | |
| *** shinylasers has joined #openstack-dev | 18:05 | |
| *** nati_uen_ has quit IRC | 18:05 | |
| *** ffio has quit IRC | 18:05 | |
| *** shinylasers has quit IRC | 18:05 | |
| *** nati_ueno has quit IRC | 18:05 | |
| *** nati_uen_ has joined #openstack-dev | 18:05 | |
| *** d0ugal has joined #openstack-dev | 18:06 | |
| *** shinylasers has joined #openstack-dev | 18:06 | |
| *** sushils has quit IRC | 18:06 | |
| *** shinylasers has quit IRC | 18:06 | |
| *** d0ugal has quit IRC | 18:06 | |
| *** d0ugal has joined #openstack-dev | 18:07 | |
| *** d0ugal has quit IRC | 18:07 | |
| *** jcooley_ has joined #openstack-dev | 18:08 | |
| *** d0ugal has joined #openstack-dev | 18:08 | |
| *** d0ugal has joined #openstack-dev | 18:08 | |
| *** d0ugal has quit IRC | 18:08 | |
| tellesnobrega | dolphm: is there a way that i can limit actions on the policy json by the domain, i tried domain_id=%(domain_id)s and i authenticated with a domain scoped token, but it doesnt work | 18:09 |
| *** d0ugal has joined #openstack-dev | 18:10 | |
| *** d0ugal has quit IRC | 18:10 | |
| *** d0ugal has joined #openstack-dev | 18:10 | |
| *** d0ugal has quit IRC | 18:10 | |
| *** gothicmindfood has joined #openstack-dev | 18:11 | |
| jamielennox | tellesnobrega: have a look at the policy.v3cloudsample.json for some examples | 18:12 |
| tellesnobrega | jamielennox: thanks | 18:12 |
| jamielennox | you will need to set a role as well, as all that will check on it's own is that there is a domain | 18:12 |
| *** shinylasers has joined #openstack-dev | 18:12 | |
| *** markwash has joined #openstack-dev | 18:12 | |
| *** shinylasers has quit IRC | 18:13 | |
| *** davidhadas has joined #openstack-dev | 18:13 | |
| *** d0ugal has joined #openstack-dev | 18:13 | |
| *** d0ugal has joined #openstack-dev | 18:13 | |
| *** d0ugal has quit IRC | 18:13 | |
| *** jcooley_ has quit IRC | 18:13 | |
| jamielennox | ayoung has a blog post recently with some fairly detailed examples, adam.younglogic.com and have a look back trhough the history | 18:13 |
| *** shinylasers has joined #openstack-dev | 18:13 | |
| *** shinylasers has quit IRC | 18:13 | |
| ayoung | tellesnobrega, http://adam.younglogic.com/2013/11/policy-enforcement-in-openstack/ | 18:14 |
| tellesnobrega | jamielennox: i have a role that i created and setted the domain for it | 18:14 |
| tellesnobrega | ayoung: i've already used this post to guide through the first tests | 18:15 |
| tellesnobrega | it worked nicely here, but my problem is about the domain, how to check if the user is owner of the domain on context to allow it to perform actions | 18:15 |
| *** jkyle_ is now known as jkyle | 18:16 | |
| *** jkyle is now known as Guest75169 | 18:16 | |
| rpodolyaka1 | dhellmann: ping | 18:17 |
| dhellmann | rpodolyaka1: in a meeting, email me? | 18:17 |
| *** dhellmann is now known as dhellmann_ | 18:17 | |
| rpodolyaka1 | dhellmann_: ack | 18:17 |
| *** jtomasek has quit IRC | 18:17 | |
| *** rongze has quit IRC | 18:18 | |
| *** Guest75169 has quit IRC | 18:18 | |
| jamielennox | sdague: i'm consistently failing grenade with this error: http://logs.openstack.org/01/59601/10/check/check-grenade-dsvm/bd0d43b/logs/new/screen-s-proxy.txt.gz | 18:18 |
| jamielennox | sdague: can you have a look and tell me if it's something you know about? | 18:19 |
| *** d0ugal has joined #openstack-dev | 18:19 | |
| *** rfolco has quit IRC | 18:20 | |
| *** rfolco has joined #openstack-dev | 18:21 | |
| *** SumitNaiksatam has joined #openstack-dev | 18:21 | |
| *** pdevine has quit IRC | 18:22 | |
| *** athomas has quit IRC | 18:22 | |
| *** harlowja_away is now known as harlowja | 18:22 | |
| *** d0ugal has joined #openstack-dev | 18:22 | |
| *** d0ugal has quit IRC | 18:22 | |
| *** d0ugal has joined #openstack-dev | 18:22 | |
| sdake_ | giulivo tempest meeting is not here https://wiki.openstack.org/wiki/Meetings - do you know the times? | 18:23 |
| giulivo | sdague, sure, i added the topic in the agenda | 18:24 |
| *** d0ugal has joined #openstack-dev | 18:24 | |
| *** d0ugal has quit IRC | 18:24 | |
| giulivo | https://wiki.openstack.org/wiki/Meetings/QATeamMeeting | 18:24 |
| giulivo | sdake, ^^ sorry | 18:24 |
| *** mnaser has quit IRC | 18:24 | |
| *** davidhadas_ has joined #openstack-dev | 18:24 | |
| sdake_ | oh thanks - :) | 18:24 |
| giulivo | sdake, it's referenced as QA team meeting not Tempest team meeting | 18:25 |
| *** vipul is now known as vipul-away | 18:26 | |
| *** shinylasers has joined #openstack-dev | 18:26 | |
| *** shinylasers has quit IRC | 18:26 | |
| *** abramley has joined #openstack-dev | 18:26 | |
| *** d0ugal has joined #openstack-dev | 18:26 | |
| *** mnaser has joined #openstack-dev | 18:26 | |
| *** d0ugal has joined #openstack-dev | 18:27 | |
| *** d0ugal has quit IRC | 18:27 | |
| *** d0ugal has joined #openstack-dev | 18:27 | |
| sdake_ | giulivo reading scrollback, I think the issue is that we would prefer tests to be tracked in blueprints from a downstream perspective | 18:27 |
| *** shinylasers has joined #openstack-dev | 18:27 | |
| *** d0ugal has joined #openstack-dev | 18:27 | |
| *** d0ugal has quit IRC | 18:27 | |
| *** d0ugal has joined #openstack-dev | 18:27 | |
| sdake_ | etherpads change, items get deleted or added, whether we like launchpad or not, we are stuck with it :) | 18:27 |
| sdake_ | it creates a permanent record of changes, which is what downstreams want | 18:28 |
| *** shinylasers has joined #openstack-dev | 18:28 | |
| *** davidhadas has quit IRC | 18:28 | |
| *** shinylasers has quit IRC | 18:28 | |
| *** shinylasers has joined #openstack-dev | 18:28 | |
| sdake_ | all openstack projects run using launchpad and blueprints - test = new feature = blueprint | 18:29 |
| *** blamar has quit IRC | 18:29 | |
| *** shinylasers has quit IRC | 18:29 | |
| *** d0ugal has joined #openstack-dev | 18:29 | |
| sdake_ | while I'm certain the idea of using etherpads didn't evolve out of a vacuum, just trying to understand the rationale, because it doesn't fit with what I would have expected | 18:29 |
| *** shinylasers has joined #openstack-dev | 18:29 | |
| *** shinylasers has quit IRC | 18:29 | |
| *** shinylasers has joined #openstack-dev | 18:30 | |
| *** shinylasers has quit IRC | 18:30 | |
| *** shinylasers has joined #openstack-dev | 18:31 | |
| *** shinylasers has quit IRC | 18:31 | |
| *** tdruiva_ has joined #openstack-dev | 18:32 | |
| *** shinylasers has joined #openstack-dev | 18:32 | |
| *** shinylasers has quit IRC | 18:32 | |
| *** MaxV has joined #openstack-dev | 18:33 | |
| *** shinylasers has joined #openstack-dev | 18:33 | |
| *** tdruiva_ has quit IRC | 18:35 | |
| *** arozumenko1 has joined #openstack-dev | 18:35 | |
| *** tdruiva has quit IRC | 18:35 | |
| *** d0ugal has joined #openstack-dev | 18:36 | |
| *** d0ugal has quit IRC | 18:36 | |
| *** d0ugal has joined #openstack-dev | 18:36 | |
| *** tdruiva has joined #openstack-dev | 18:36 | |
| giulivo | sdake, I think the etherpad should only cover the "assignments" issue | 18:36 |
| *** eglynn has joined #openstack-dev | 18:36 | |
| *** vkozhukalov has quit IRC | 18:36 | |
| *** d0ugal has joined #openstack-dev | 18:36 | |
| *** arozumenko has quit IRC | 18:36 | |
| *** tan has joined #openstack-dev | 18:37 | |
| *** d0ugal has joined #openstack-dev | 18:37 | |
| *** d0ugal has quit IRC | 18:37 | |
| *** d0ugal has joined #openstack-dev | 18:37 | |
| sdake_ | giulivo I see you have added something for the 19th, i'll make sure to attend to add my viewpoint as a heat-core dev | 18:37 |
| *** vkozhukalov has joined #openstack-dev | 18:37 | |
| *** ifarkas has joined #openstack-dev | 18:37 | |
| giulivo | sdake, indeed | 18:38 |
| sdake_ | i gotta run dr appointment thanks! | 18:38 |
| *** hartsocks has joined #openstack-dev | 18:38 | |
| *** Abhishek_ has quit IRC | 18:38 | |
| *** blamar has joined #openstack-dev | 18:39 | |
| *** d0ugal has joined #openstack-dev | 18:40 | |
| *** d0ugal has joined #openstack-dev | 18:40 | |
| *** d0ugal has quit IRC | 18:40 | |
| *** athomas has joined #openstack-dev | 18:40 | |
| *** alexpilotti has quit IRC | 18:42 | |
| *** jcoufal has joined #openstack-dev | 18:42 | |
| *** d0ugal has joined #openstack-dev | 18:42 | |
| *** d0ugal has quit IRC | 18:42 | |
| *** d0ugal has joined #openstack-dev | 18:42 | |
| *** jistr has joined #openstack-dev | 18:44 | |
| *** sarob has joined #openstack-dev | 18:44 | |
| *** jcooley_ has joined #openstack-dev | 18:44 | |
| *** jcooley_ has quit IRC | 18:44 | |
| *** jcooley_ has joined #openstack-dev | 18:45 | |
| *** d0ugal has joined #openstack-dev | 18:45 | |
| *** d0ugal has quit IRC | 18:45 | |
| *** d0ugal has joined #openstack-dev | 18:45 | |
| *** jcooley_ has quit IRC | 18:46 | |
| *** jcooley_ has joined #openstack-dev | 18:46 | |
| *** vipul-away is now known as vipul | 18:46 | |
| *** bcrochet has quit IRC | 18:47 | |
| *** d0ugal has joined #openstack-dev | 18:47 | |
| *** d0ugal has quit IRC | 18:47 | |
| *** d0ugal has joined #openstack-dev | 18:47 | |
| *** d0ugal has joined #openstack-dev | 18:48 | |
| *** d0ugal has joined #openstack-dev | 18:48 | |
| *** d0ugal has quit IRC | 18:48 | |
| *** garyk has quit IRC | 18:49 | |
| *** garyk has joined #openstack-dev | 18:49 | |
| *** kushal has quit IRC | 18:49 | |
| *** d0ugal has joined #openstack-dev | 18:50 | |
| *** Drankis has quit IRC | 18:50 | |
| *** johnthetubaguy1 has quit IRC | 18:50 | |
| *** jcooley_ has quit IRC | 18:50 | |
| *** d0ugal has quit IRC | 18:51 | |
| *** jcooley_ has joined #openstack-dev | 18:51 | |
| *** lucasagomes has quit IRC | 18:51 | |
| *** MaxV has quit IRC | 18:51 | |
| *** d0ugal has joined #openstack-dev | 18:52 | |
| *** MaxV has joined #openstack-dev | 18:52 | |
| *** mikeoutland has quit IRC | 18:52 | |
| *** johnthetubaguy has joined #openstack-dev | 18:53 | |
| *** prekarat has quit IRC | 18:53 | |
| *** MaxV has quit IRC | 18:53 | |
| *** nati_ueno has joined #openstack-dev | 18:53 | |
| *** d0ugal has quit IRC | 18:54 | |
| *** d0ugal has joined #openstack-dev | 18:54 | |
| *** bcrochet has joined #openstack-dev | 18:54 | |
| *** MaxV has joined #openstack-dev | 18:54 | |
| *** jcooley_ has quit IRC | 18:55 | |
| *** spzala has quit IRC | 18:56 | |
| *** Mandell has joined #openstack-dev | 18:56 | |
| *** nati_uen_ has quit IRC | 18:57 | |
| *** MaxV has quit IRC | 18:58 | |
| *** rushiagr has quit IRC | 18:59 | |
| *** Alexei_987 has quit IRC | 18:59 | |
| *** vipul is now known as vipul-away | 19:01 | |
| ayoung | dolphm, Here's the proposal: | 19:01 |
| ayoung | If the client is trying to do v3, and the url ends in v2.0, chop if off and do discovery | 19:02 |
| *** fabiog has joined #openstack-dev | 19:02 | |
| ayoung | We can't do less than that | 19:02 |
| fabiog | morganfainberg: ping | 19:02 |
| *** jtomasek has joined #openstack-dev | 19:02 | |
| *** fabiog has quit IRC | 19:02 | |
| morganfainberg | dstanek, ok, so in the refactor for the controller(s), which logic seemed to be missing frmo the delte_user / delete_group etc? I was trying to limit change in logic in the same space i refactored where the logic occured | 19:03 |
| dolphm | atiwari: your example illustrates redirects, which means the client is involved; it's the same pattern as illustrated in the other two diagrams mentioned during the meeting | 19:03 |
| marekd | morganfainberg: i agree with you and ECP. I don't see any value in having federation via SAML only with browsers. | 19:03 |
| jamielennox | fabiog: have a link handy? | 19:03 |
| morganfainberg | dstanek, i'm happy to change the logic, but want to make sure we're headed in a good direction :) | 19:03 |
| *** fabio_ has joined #openstack-dev | 19:03 | |
| *** fabio_ is now known as fabiog | 19:03 | |
| atiwari | dolphm, we can avoid it for now as per summit it was not the focus for now | 19:03 |
| *** thomasm_ has joined #openstack-dev | 19:03 | |
| morganfainberg | fabiog, Let me know if you have any questions about my comment i just posted (during the meeting) | 19:03 |
| dstanek | morganfainberg: the only thing that's different is that one has a try-except right? | 19:04 |
| dolphm | atiwari: yeah, that specifically example isn't part of "use case 1", right? | 19:04 |
| ayoung | the other things I was suggesting, such as appending v3 on to the URL, while ugly, reflet the reality that we don't have a caching strategy in place, and so we can't remember the results of previous discovery | 19:04 |
| ayoung | So I think they fall into "necessary performance hacks" but we can punt on them to start | 19:04 |
| dolphm | ayoung: so, implement a caching strategy? | 19:04 |
| morganfainberg | dstanek, correct. since deletes look up the ref and that would fail and fail the deletion | 19:04 |
| atiwari | dolphm, yes. let not discuss it | 19:04 |
| *** jcooley_ has joined #openstack-dev | 19:04 | |
| morganfainberg | dstanek, figured it was sane to just move on and not explode. | 19:04 |
| *** vartom111119 has joined #openstack-dev | 19:05 | |
| ayoung | dolphm, all of that is performance tune. | 19:05 |
| ayoung | so, yes | 19:05 |
| ayoung | but we need to have the v3 api deal with getting a v2.0 url | 19:05 |
| morganfainberg | dstanek, and i didn't want to move the delete_token* and credential deletes to the manager, imo the manager shouldn't cross talk to other managers unless it's really needed | 19:05 |
| *** salv-orlando has quit IRC | 19:05 | |
| morganfainberg | dstanek, it seems that manager crossing is more of a "controller" business logic thing | 19:05 |
| *** Drankis has joined #openstack-dev | 19:05 | |
| dstanek | morganfainberg: no what i was saying is that you that duplication in the controller itself now | 19:06 |
| dstanek | morganfainberg: i agree that since deletes are idempotent that the try-except is a good thing - so why is it not in the delete_* methods? | 19:06 |
| dstanek | morganfainberg: this if that's the case the code is exactly the same | 19:07 |
| morganfainberg | dstanek, oh i see, move the try/excepts down | 19:07 |
| *** thomasem has quit IRC | 19:07 | |
| fabiog | morganfainberg, a standard object method can be registered n times | 19:07 |
| *** Nikolay_St has joined #openstack-dev | 19:07 | |
| fabiog | morganfainberg, if we go this route we put the burden in the extension code | 19:07 |
| fabiog | which is fine with me | 19:07 |
| morganfainberg | fabiog, correct, but @classmethod can't reference (let's use a bad example, oauth consumers) oauth_api.delete_consumer | 19:07 |
| morganfainberg | fabiog, since that is a bound instance method. and likely on a user delete you'd want to cleanup consumers | 19:08 |
| fabiog | merganfainberg, not sure I am following you | 19:08 |
| fabiog | merganfainberg, right | 19:08 |
| dstanek | morganfainberg: if you moved it down then you can just call the delete_* methods from the for loop | 19:08 |
| *** jcooley_ has quit IRC | 19:08 | |
| morganfainberg | dstanek, ah. yes i agree, i think that is a massive change though (and likely affects tempest and a ton of tests that rely on delete returning an error if things don't exist) | 19:09 |
| dstanek | morganfainberg: also i noticed that the logic in the "delete group" for loop is not the same as delete_group() | 19:09 |
| morganfainberg | dstanek, i'm happy to tackle that aspect separately | 19:09 |
| morganfainberg | dstanek, let me check what i changed there. | 19:09 |
| *** nati_ueno has quit IRC | 19:10 | |
| *** danjared_ has quit IRC | 19:10 | |
| fabiog | merganfainberg, can't you call an instance of the Driver from there to take care of that? | 19:10 |
| dstanek | morganfainberg: does that mean the having the change in _delete_domain_contents() may also break tempest? | 19:10 |
| morganfainberg | dstanek, ah, i think we catch that above since groups don't cross domains? | 19:10 |
| morganfainberg | dstanek, on line 603 | 19:10 |
| morganfainberg | dstanek, if you think it is an issue i can add the users from a group to the "user_refs" (make it a set instead) | 19:11 |
| *** zaitcev has joined #openstack-dev | 19:11 | |
| *** hajay_ has left #openstack-dev | 19:11 | |
| morganfainberg | around like 611, i don't want to loop through 3 user lists. we already do 2 | 19:11 |
| *** dvarga has quit IRC | 19:11 | |
| *** mikeoutland has joined #openstack-dev | 19:11 | |
| morganfainberg | and the domain lookup logic in delete_group is superceeded by r['domain_id'] | 19:12 |
| morganfainberg | dstanek, i think we're safe on that one, since we really didn't test broken deletes | 19:12 |
| *** melwitt has joined #openstack-dev | 19:12 | |
| *** rongze has joined #openstack-dev | 19:13 | |
| *** nati_ueno has joined #openstack-dev | 19:13 | |
| shardy | ayoung: Hi | 19:13 |
| ayoung | shardy, wazzup | 19:13 |
| *** Mandell has quit IRC | 19:13 | |
| morganfainberg | fabiog, @classmethods are created (effectively) at definition time. how do you call an instance of a driver from a non-instance-bound method, 'cls' doesn't have the reference. i'm trying to make sure we aren't blueprinting issues into our examples here | 19:13 |
| atiwari | guys I need some blessing on https://review.openstack.org/#/c/61897, whenever have time. | 19:13 |
| shardy | ayoung: Got a moment to follow up on the x509/oauth discussion for in-instance users? | 19:13 |
| *** Mandell has joined #openstack-dev | 19:14 | |
| ayoung | shardy, oauth? | 19:14 |
| ayoung | Sure | 19:14 |
| dstanek | morganfainberg: will that code on line 603 include all users of the groups being deleted? | 19:14 |
| morganfainberg | dstanek, if groups don't bridge domains, yes | 19:15 |
| shardy | ayoung: I'm trying to figure out what the roadmap is as there seem to be several aspects missing for oauth (client support, signature generation/validation, auth middleware) | 19:15 |
| morganfainberg | dstanek, if they do, i think we have other issues | 19:15 |
| ayoung | shardy, I don't think you want oauth | 19:15 |
| morganfainberg | dstanek, and this whole bit of code should be scrapped. | 19:15 |
| shardy | ayoung: re your x509 comment, is that some future thing, or is it ready now? | 19:15 |
| morganfainberg | dstanek, i don't see how a group could bridge an IdP (sorry not domain, IdP) | 19:15 |
| *** dolphm has quit IRC | 19:16 | |
| shardy | ayoung: Ok, dolphm indicated the opposite, hence myself (and therve) have been trying to figure it out :) | 19:16 |
| morganfainberg | dstanek, (scrub "domain" in relation to group and replace with IdP where possible) | 19:16 |
| *** herndon has quit IRC | 19:16 | |
| ayoung | shardy, x509 can be used now, but it is not a general purpose solution. It implieas a degree of infrastructure that the Heat team may not be able to count on being available | 19:16 |
| *** dvarga has joined #openstack-dev | 19:16 | |
| ayoung | yes, I know, but oauth, according to the spec and the implementation, is short term delegation | 19:16 |
| shardy | ayoung: Ok, for now, I'm working on using a trust, to delegate to a "proxy user" who will create an ec2 keypair with a trust scoped token | 19:16 |
| *** rushiagr has joined #openstack-dev | 19:16 | |
| shardy | then the keypair can be used to authenticate, impersonating the user owning the heat stack, but with minimal roles | 19:17 |
| shardy | ayoung: but some users don't want ec2tokens | 19:17 |
| shardy | so I'm trying to figure out he alternative | 19:17 |
| ayoung | shardy, the root problem, as I understand it, is that we have nothing in Keystone to represent the entity to whom you need to delegate | 19:17 |
| *** sarob has quit IRC | 19:17 | |
| shardy | and I also don't really like the reliance on the proxy user | 19:17 |
| ayoung | its basically the same problem as KDS...we have not IdM for all of the undercloud | 19:18 |
| *** rongze has quit IRC | 19:18 | |
| shardy | ayoung: right, so I'm creating a heat_proxy_user in a heat_proxy tenant, who has no roles | 19:18 |
| ayoung | so you need a long term login mechanism for that user. | 19:18 |
| shardy | just so I can delegate to the proxy so I can leverage trusts to role-limit access | 19:18 |
| ayoung | Don't misue trusts to do that | 19:18 |
| *** gyee has quit IRC | 19:18 | |
| shardy | ayoung: Well AFAICS there is no alternative? | 19:19 |
| morganfainberg | fabiog, i think we need put the burden on the extension code to "do the right thing" in registration etc. if there is nothing to do (because a previous instance acted on the notification) it should be easy to skip. | 19:19 |
| ayoung | ec2creds made the same mistake | 19:19 |
| shardy | ayoung: the alternative is every user who creates a heat stack has to be admin | 19:19 |
| morganfainberg | fabiog, i'm fine with removing my -1, just wanted to make sure we thought this through before pushing it in | 19:19 |
| shardy | because currently we create a new user to own every ec2 keypair we put in an instance | 19:19 |
| ayoung | no | 19:19 |
| *** gothicmindfood has quit IRC | 19:19 | |
| ayoung | the alkternative is that the heat user gets multiple trusts set up | 19:20 |
| ayoung | and you need to do password sync | 19:20 |
| *** herndon has joined #openstack-dev | 19:20 | |
| dstanek | morganfainberg: looks like the sql backend enforces a group in be within a domain so i think that filter is enough - is that a requirement of backends? | 19:20 |
| ayoung | I *think* yo uare trying to avoid giving too much power to one user | 19:20 |
| ayoung | which is a noble effort | 19:20 |
| *** jcoufal-mob has joined #openstack-dev | 19:20 | |
| fabiog | morganfainberg, you have a valid point | 19:20 |
| *** danjared has joined #openstack-dev | 19:20 | |
| shardy | ayoung: I'm trying to limit what the credentials we put in the (implicitly untrusted) instance can do | 19:20 |
| shardy | ayoung: but still enable authentication with out API | 19:21 |
| morganfainberg | dstanek, with LDAP i can confirm it's the same, since LDAP is non-domain aware | 19:21 |
| shardy | ayoung: normal (non admin) users must be able to use whatever method we provide, so trusts is a good starting point | 19:21 |
| fabiog | morganfainberg, I am trying to make sure that the solution is robust enough to avoid bad practices and at the same time re-use the Drivers to put the data in sync | 19:21 |
| morganfainberg | dstanek, i think it is a fair-assessment that it is a requirement (though not hard-enforced anywhere but the backend) | 19:21 |
| shardy | but the reliance on ec2-keypairs is a problem | 19:21 |
| *** jcoufal-mob has quit IRC | 19:22 | |
| morganfainberg | fabiog, i totally understand, which is why i wanted to discuss it instead of -1 and let you figure it out :) | 19:22 |
| ayoung | shardy, you are asking me a hard problem. I don't have an easy answer for you. | 19:22 |
| shardy | ayoung: One observation is, if we made token expiry optional for trust scoped tokens, all our problems would be solved | 19:22 |
| *** jcoufal-mob has joined #openstack-dev | 19:22 | |
| shardy | ayoung: Is that even a possibility? | 19:22 |
| ayoung | shardy, no | 19:22 |
| morganfainberg | fabiog, it warrants a think through vs a "nope, i don't like it" | 19:22 |
| ayoung | no no no noi no noi no no non negatory | 19:22 |
| ayoung | shardy, the question is "how do I delegate to an ephemeral user" | 19:23 |
| morganfainberg | ayoung, +++ no non-expiring bearer tokens please | 19:23 |
| shardy | ayoung: Well that is all we're doing, by proxy, by using other non-expiring credentials which can be used to obtain a token | 19:23 |
| fabiog | morgainfainberg, do you think that introducing a singleton class that is not the controllers would be a better approach? | 19:23 |
| *** urulama has joined #openstack-dev | 19:23 | |
| *** jcooley_ has joined #openstack-dev | 19:23 | |
| shardy | only it introduces a lot of pain because it's a non-core extension (ec2tokens) and doesn't use the standard auth_token middleware | 19:23 |
| *** ruhe has joined #openstack-dev | 19:23 | |
| *** jcoufal has quit IRC | 19:23 | |
| morganfainberg | fabiog, if the controllers were made singletons - it might be easier. | 19:24 |
| shardy | ayoung: I agree, this is a hard problem, but it's existed since heat started, and I'm making a real effort to understand how to solve it | 19:24 |
| ayoung | shardy, tokens are authorization documents, not authentication. I said X509 because it is the industry standard for that. It still would require a user to be authenticated as, though, in order for the trust to work. Oauth avoids some of that by using consuemrs instreadof users, but so what | 19:24 |
| morganfainberg | fabiog, but i don't know if that is the best approach either. | 19:24 |
| *** johnthetubaguy has quit IRC | 19:24 | |
| fabiog | morganfainberg, no I am talking about a new class, like notification handler | 19:24 |
| fabiog | which has the callbacks | 19:24 |
| *** troytoman-away is now known as troytoman | 19:25 | |
| morganfainberg | fabiog, oh so you'd ... map a callback instance to all "controllers" of a type? | 19:25 |
| *** xqueralt has joined #openstack-dev | 19:25 | |
| ayoung | shardy, I think the right answer is to havea heat domain, and each instance is a new user, and the heat manager should beable to create heat users that have no roles, to consume trusts. These users would be authenticated via X509 if possibble, but symmetric shared secrets (authgened passwords) is the fallback for the trivial case | 19:25 |
| morganfainberg | callback logic singleton that is? | 19:25 |
| *** jcooley_ has quit IRC | 19:25 | |
| ayoung | by "instance" I mean | 19:25 |
| fabiog | morganfainberg, more of an extension. | 19:26 |
| ayoung | the heat proxy user | 19:26 |
| ayoung | not one for all users, but one per workflow | 19:26 |
| ayoung | we call them users, but really they are nothing more than placeholder accounts | 19:26 |
| shardy | ayoung: interesting, so that is essentially an extension to what I'm doing, where we replace the fixed "proxy user" with the per-instance user in the heat domain | 19:26 |
| *** Mandell has quit IRC | 19:26 | |
| fabiog | morganfainberg, because if the extension registers several times there could be really nasty effects and it is going to be hard to debug | 19:26 |
| morganfainberg | fabiog, by using inheritance or a metaclass? or? | 19:26 |
| ayoung | used to minimize damage if one should get compromised | 19:26 |
| *** MaxV has joined #openstack-dev | 19:26 | |
| *** yolanda has quit IRC | 19:26 | |
| shardy | ayoung: I like that idea, as although it still means the overhead of creating lots of users in keystone, they won't be visible to users in their domain | 19:27 |
| *** urulama has quit IRC | 19:27 | |
| morganfainberg | fabiog, remember the issue still occurs that the singleton needs to be aware of the instantiated objects (and the like) | 19:27 |
| shardy | ayoung: I assume trust delegation/impersonation can work accross domains? | 19:27 |
| ayoung | shardy, yes. POKI is better than symmetric here, and X509 is the best method to do PKI, but it works with lesser methods | 19:27 |
| ayoung | yes, trusts is cross domina | 19:27 |
| ayoung | domain | 19:27 |
| morganfainberg | fabiog, or are you advocating a completely new class that gets all the dependancies added etc (@dependancy) that has logic to handle the callback? | 19:28 |
| shardy | ayoung: This is very cool, thanks for working through this, what you suggest may work well for us :) | 19:28 |
| *** rpodolyaka1 has left #openstack-dev | 19:28 | |
| morganfainberg | fabiog, sorry i'm tracking a few conversations, so i need it spelled out somtimes :P | 19:28 |
| fabiog | morganfainberg, yes, this is what I am advocating | 19:28 |
| ayoung | shardy, glad to hear it | 19:28 |
| *** jasondotstar is now known as jclark_ | 19:28 | |
| *** jcooley_ has joined #openstack-dev | 19:28 | |
| *** jclark_ is now known as jasondotstar | 19:28 | |
| shardy | ayoung: Can I write up what we've discussed in a wiki page, and ask you to review it please? | 19:29 |
| ayoung | shardy, yep | 19:29 |
| morganfainberg | fabiog, i like that approach better. | 19:29 |
| *** sarob has joined #openstack-dev | 19:29 | |
| shardy | ayoung: Thanks, I'll ping you a link later! | 19:29 |
| fabiog | morganfainberg, if there is a way of making that class a singleton and using the @dependency inject the Driver we should be good | 19:29 |
| fabiog | morganfainberg, do you think that is feasible? | 19:29 |
| ayoung | fabiog, it is rare that you want a true singleton | 19:29 |
| *** thomasem has joined #openstack-dev | 19:29 | |
| morganfainberg | ayoung, in this case, i think he does. | 19:29 |
| ayoung | waht you want is a known named instance of a class, and the current dep injection does that | 19:30 |
| ayoung | morganfainberg, I doubt it | 19:30 |
| *** Mandell has joined #openstack-dev | 19:30 | |
| morganfainberg | ayoung, a single object that doesn't get re-instantiated (ever) to handle the callbacks on events for extenstion notifications | 19:30 |
| *** harlowja is now known as harlowja_away | 19:30 | |
| *** marekd is now known as marekd|away | 19:30 | |
| ayoung | morganfainberg, one per extension? | 19:30 |
| morganfainberg | ayoung, though it doesn't need to be a "singleton" in mechanics, just in use (duck-typed languages and all) | 19:30 |
| stevebaker | ttx: ping? | 19:30 |
| ttx | stevebaker: hai! | 19:30 |
| *** jcooley_ has quit IRC | 19:31 | |
| *** jcooley_ has joined #openstack-dev | 19:31 | |
| ttx | stevebaker: https://launchpad.net/heat/+milestone/icehouse-2 | 19:31 |
| *** romcheg has joined #openstack-dev | 19:31 | |
| morganfainberg | fabiog, i think if you use something akin to the @dependancy.provides mechanism, that should be sufficient to avoid issues with multiple instantiations | 19:31 |
| ttx | stevebaker: only one to set priority for: https://blueprints.launchpad.net/heat/+spec/add-parameter-label-to-template | 19:31 |
| ttx | and 4 missing assignees | 19:32 |
| ttx | err.. make that 5 | 19:32 |
| morganfainberg | ayoung, yes, one per extension, since hte logic would be different per extension | 19:32 |
| fabiog | morganfainberg, so I don't need a new class, I can just use that in the current controllers | 19:32 |
| ttx | stevebaker: made progress in finding assignees for them ? | 19:32 |
| stevebaker | ttx: that is in discussion, it may be rejected http://ask-openstackux.rhcloud.com/question/119/should-heat-pass-parameter-labels-to-horizon/ | 19:32 |
| *** thomasm_ has quit IRC | 19:32 | |
| ttx | stevebaker: ok, keep in undefined until you know | 19:33 |
| morganfainberg | fabiog, hm. i think you could do it with the @dependancy mechanism | 19:33 |
| *** colinmcnamara has joined #openstack-dev | 19:33 | |
| stevebaker | ttx: I asked for volunteers, maybe one blueprint was assigned from that | 19:33 |
| *** colinmcnamara has quit IRC | 19:33 | |
| morganfainberg | fabiog, you likely don't need to involve the controllers at all | 19:34 |
| ttx | stevebaker: I think if nobody is signed up for them at his point you might consider removing from the milestone | 19:34 |
| morganfainberg | fabiog, if you used @dependancy.provides to call a "register_subscriptions" the manager could do the callback for youy | 19:34 |
| stevebaker | ttx: agreed | 19:34 |
| morganfainberg | fabiog, the controller doesn't need to be involved at all. | 19:34 |
| ttx | stevebaker: or maybe drop them to "Low" prio and give it one more week | 19:34 |
| *** eglynn has quit IRC | 19:35 | |
| ttx | stevebaker: everything else, including progress, looks good to me | 19:35 |
| stevebaker | ttx: I'll go through each one and make a call | 19:35 |
| ttx | stevebaker: we'll skip the next 2 1:1s and cross-project meetings | 19:35 |
| *** e0ne has joined #openstack-dev | 19:35 | |
| ttx | (on te 24 and 31) | 19:35 |
| *** rushiagr has quit IRC | 19:36 | |
| morganfainberg | fabiog, so you get the "<dependancy>_api" as the listener. not a true singleton, but the registration code isn't called externally on subsequent instances. ( cc ayoung, no singleton needed) | 19:36 |
| ttx | stevebaker: do you have anything specific you want to see discussed at the meeting later ? | 19:36 |
| stevebaker | ttx: actually I won't be back until Jan 6 | 19:36 |
| ayoung | morganfainberg, yep, good enough | 19:36 |
| *** markwash has quit IRC | 19:36 | |
| ttx | stevebaker: awesome, enjoy | 19:36 |
| dstanek | morganfainberg: i think i have to come back and look at that more - feels wrong to have a refactor add duplication | 19:37 |
| fabiog | morganfainberg, let me see if I got it: I register methods that are in the ExampleManager, which are under the dependency.provider and I am done? | 19:37 |
| stevebaker | ttx: I'll find out tomorrow if we skip 2 weeks of heat meetings, or just one. I'll let you know if there is a fill-in ptl to delegate to for the first week of jan | 19:37 |
| morganfainberg | dstanek, i know. it's actually reducing some duplication. | 19:37 |
| ttx | stevebaker: next meeting is on the 7th, so you should be back | 19:38 |
| morganfainberg | dstanek, it doesn't feel like it. the credential_api code was icky so i pushed that code down and just called managers directly | 19:38 |
| *** dave_tucker_ has joined #openstack-dev | 19:38 | |
| stevebaker | ttx: ok, cool | 19:39 |
| morganfainberg | dstanek, i have one more patchset in that chain but it's a doozy so i want to get these two done. i can include further de-dup and idempotent-awesomeness at that point. heck i think a chunk of that code cleaning out the domain can be removed as well since some bugs were fixed | 19:39 |
| ttx | stevebaker: see you at meeting in 80min | 19:39 |
| dstanek | morganfainberg: i like how you moved the credentials methods out - i'm just struggling with the delete changes | 19:39 |
| fabiog | morganfainberg, something like: core.ExampleManager.project_created_notification | 19:39 |
| morganfainberg | fabiog, well you'll want to do it in the section of the code that handles the dependancy and making sure it's instantiated | 19:40 |
| *** utlemming has quit IRC | 19:40 | |
| morganfainberg | but yes | 19:40 |
| *** utlemming has joined #openstack-dev | 19:40 | |
| stevebaker | ttx: ack | 19:40 |
| *** sstent has quit IRC | 19:40 | |
| *** epim has joined #openstack-dev | 19:41 | |
| morganfainberg | dstanek, if you think about it from the standpoint of removing instantiation of a controller to call delete, and calling the logic directly, it makes it a bit more palatable | 19:41 |
| fabiog | morganfainberg, ok. will do the changes later today. Thanks | 19:41 |
| *** ayoung is now known as ayoung_xc-ski | 19:41 | |
| *** rollietikes has quit IRC | 19:41 | |
| *** colinmcnamara has joined #openstack-dev | 19:41 | |
| *** arozumenko1 has quit IRC | 19:42 | |
| morganfainberg | dstanek, it was doing a pretty heavy process to de-dupe code. | 19:42 |
| *** sushils has joined #openstack-dev | 19:42 | |
| morganfainberg | dstanek, let me know what you decide so i can work on either correcting it or slating more cleanup in the chain :) | 19:44 |
| dstanek | morganfainberg: in your opinion where does that logic belong? | 19:44 |
| *** dprince has quit IRC | 19:44 | |
| *** CaptTofu has quit IRC | 19:45 | |
| morganfainberg | dstanek, in my opinion controllers shouldn't instantiate eachother for logic. it should be in the manager apis | 19:45 |
| *** dprince has joined #openstack-dev | 19:45 | |
| morganfainberg | dstanek, this is an odd case because domain cleanup is ... odd | 19:45 |
| *** CaptTofu has joined #openstack-dev | 19:46 | |
| dstanek | morganfainberg: do the managers typically call other managers? | 19:46 |
| morganfainberg | dstanek, no. | 19:46 |
| morganfainberg | dstanek, well... in a couple cases yes | 19:47 |
| *** nermina has quit IRC | 19:47 | |
| therve | shardy, I'm very interested in that wiki page as well :) | 19:47 |
| morganfainberg | e.g. token, token_provider | 19:47 |
| *** dave_tucker_ has quit IRC | 19:48 | |
| morganfainberg | and assignment -> identity | 19:48 |
| morganfainberg | i guess we could push this down to the identity api | 19:48 |
| morganfainberg | (the delete credentials part) | 19:48 |
| *** mikeoutland has quit IRC | 19:48 | |
| morganfainberg | and we, in theory, could push the delete_tokens_for* down to the token manager | 19:48 |
| *** yassine has quit IRC | 19:49 | |
| morganfainberg | ayoung_xc-ski, bknudson, your opinion on managers calling other manager logic? Should that be more of a controller thing? (e.g. delete_credentials_for_user called from a controller or from identity_api manager on user_delete) | 19:50 |
| *** jimjiang has quit IRC | 19:50 | |
| morganfainberg | dstanek, ^ | 19:50 |
| morganfainberg | personally, i think managers should have as little cross communication as possible. | 19:50 |
| *** jimjiang has joined #openstack-dev | 19:51 | |
| morganfainberg | it should rely on the "business logic" to pull that data together, i.e. controllers | 19:51 |
| *** jcooley_ has quit IRC | 19:51 | |
| morganfainberg | otherwise it seems silly to have controllers and manager layers. | 19:51 |
| dstanek | morganfainberg: that mean that you'd be back to having controllers use other controller right? | 19:51 |
| *** jcooley_ has joined #openstack-dev | 19:51 | |
| *** vipul-away is now known as vipul | 19:51 | |
| bknudson | morganfainberg: manager makes more sense than controller | 19:51 |
| *** jcoufal-mob has quit IRC | 19:52 | |
| bknudson | controller should just be taking the request and converting it to call to manager | 19:52 |
| *** vipul-away has joined #openstack-dev | 19:52 | |
| *** jtomasek has quit IRC | 19:53 | |
| *** atiwari has quit IRC | 19:53 | |
| morganfainberg | dstanek, if we were to de-dupe that code completely, yes back to controlellers instantiating and calling controllers | 19:53 |
| dstanek | bknudson: this aligns with my understanding that the manager layer is the same as a service layer in other architectures | 19:53 |
| bknudson | call it the service layer if you like, that's what I think it's supposed to be. | 19:54 |
| *** bdpayne has quit IRC | 19:54 | |
| morganfainberg | bknudson,dstanek, maybe my cognitive dissonance is the assignment vs idenity sides | 19:54 |
| morganfainberg | perhaps credential falls more to identity side than otherwise. | 19:54 |
| dstanek | morganfainberg: or that logic of "when a user is deleted their credentials are deleted" can be pushed to the manager in one place | 19:54 |
| morganfainberg | similar to the delete_tokens_for<thing> | 19:55 |
| bknudson | morganfainberg: thing with identity is you apparently won't be able to resolve users with federation (I think) | 19:55 |
| *** epim has quit IRC | 19:55 | |
| bknudson | you won't be able to say "get me the user for this id" | 19:56 |
| morganfainberg | bknudson, actually i was thinking one step further | 19:56 |
| *** jimjiang has quit IRC | 19:56 | |
| *** jcooley_ has quit IRC | 19:56 | |
| morganfainberg | bknudson, what if identity (keystone) was made into a federated source, same limitations as other idps | 19:56 |
| morganfainberg | (heck even separate process) | 19:56 |
| *** neoXsys has quit IRC | 19:56 | |
| morganfainberg | perhaps, the right answer is to make a "clean_up_for_user" "cleanup_for_group" type method | 19:57 |
| *** yolanda has joined #openstack-dev | 19:57 | |
| morganfainberg | on the manager that can be called with just the relevant ids | 19:57 |
| morganfainberg | doesn't do any checking | 19:57 |
| bknudson | morganfainberg: this is in assignments? | 19:57 |
| *** jimjiang has joined #openstack-dev | 19:57 | |
| *** pberis has joined #openstack-dev | 19:58 | |
| *** nermina has joined #openstack-dev | 19:58 | |
| morganfainberg | bknudson, the code being discussed is domainv3 controller and the _delete_domain_contents method | 19:58 |
| bknudson | (irc messages seem to be coming in slowly for me) | 19:58 |
| morganfainberg | bknudson, same. getting bursty/delays | 19:58 |
| morganfainberg | bknudson, i think we might be headed for a netsplit | 19:58 |
| *** jcooley_ has joined #openstack-dev | 19:59 | |
| bknudson | morganfainberg: is there a domain manager? | 19:59 |
| *** MaxV has quit IRC | 19:59 | |
| morganfainberg | bknudson, no, that is part of assignment | 19:59 |
| *** dhellmann_ is now known as dhellmann | 19:59 | |
| dstanek | bknudson: we were talking about https://review.openstack.org/#/c/62197/5/keystone/identity/controllers.py | 19:59 |
| *** Alexei_987 has joined #openstack-dev | 19:59 | |
| *** colinmcnamara has quit IRC | 19:59 | |
| morganfainberg | bknudson, ^ | 19:59 |
| *** MaxV has joined #openstack-dev | 20:00 | |
| morganfainberg | actually... i think i have a solution. | 20:00 |
| *** novas0x2a|laptop has joined #openstack-dev | 20:00 | |
| bknudson | morganfainberg: so if I had to guess what domainv3._delete_domain_contents should do is it should tell the identity manager to delete and it should tell the assignment manager to delete | 20:00 |
| *** jcooley_ has quit IRC | 20:00 | |
| morganfainberg | bknudson, that was what we are currently doing. it just has some duplicated code added in because the groupv3 and userv3 controllers do similar things | 20:01 |
| *** d0ugal has quit IRC | 20:01 | |
| morganfainberg | bknudson, so you see sem-duplicated code in the different controlelrs | 20:01 |
| morganfainberg | bknudson, previously it would instantiate the controller and pass the request into that controller to do the work. | 20:02 |
| morganfainberg | which... seems broken. | 20:02 |
| morganfainberg | also makes it impossible to separate assignment controllers from identity controllers | 20:02 |
| *** jimjiang has quit IRC | 20:03 | |
| *** d0ugal has joined #openstack-dev | 20:03 | |
| *** d0ugal has quit IRC | 20:03 | |
| *** d0ugal has joined #openstack-dev | 20:03 | |
| *** d0ugal has quit IRC | 20:03 | |
| bknudson | morganfainberg: thanks for cleaning this up! | 20:03 |
| *** d0ugal has joined #openstack-dev | 20:03 | |
| *** d0ugal has quit IRC | 20:03 | |
| *** d0ugal has joined #openstack-dev | 20:03 | |
| *** MaxV has quit IRC | 20:04 | |
| morganfainberg | bknudson, i am trying to make identity and assignment as split as possible so we can support federation and the like without as much mess. similarly moving controllers and routers around to make things friendlier. | 20:04 |
| bknudson | morganfainberg: "project_cntl = ProjectV3()" -- WTFLOL | 20:04 |
| morganfainberg | bknudson, the next patchset after the suffle will be to complete the per-domain-identity patch that didn't make it into havana | 20:04 |
| *** jtomasek has joined #openstack-dev | 20:05 | |
| *** ifarkas has quit IRC | 20:05 | |
| morganfainberg | bknudson, i know. right? | 20:05 |
| *** d0ugal has joined #openstack-dev | 20:05 | |
| *** ArxCruz has quit IRC | 20:06 | |
| bknudson | morganfainberg: this is going to be a great cleanup | 20:06 |
| morganfainberg | bknudson, just don't hate me too much for the next patchset in the chain...and the use of metaclasses | 20:06 |
| *** radez_g0n3 is now known as radez | 20:07 | |
| bknudson | morganfainberg: seems like keystone shouldn't be this complicated. | 20:07 |
| bknudson | really, it doesn't do that much | 20:07 |
| morganfainberg | bknudson, i know someone who would agree with you | 20:07 |
| morganfainberg | bknudson, i am working on the cleanup slowly. | 20:07 |
| morganfainberg | bknudson, and it is getting there. | 20:07 |
| morganfainberg | bknudson, but there is a lot of cleanup (also the V2 api disappearing will help a lot) | 20:08 |
| bknudson | doing operations like deleting all the projects or all the users in a domain is not scalable anyways | 20:08 |
| morganfainberg | bknudson, i totally agree with you | 20:08 |
| morganfainberg | bknudson, as far as i am concerned, domains have 2 modes "enabled and disabled" never deleted | 20:09 |
| morganfainberg | whcih solves this "cleanup" issue | 20:09 |
| *** jimjiang has joined #openstack-dev | 20:09 | |
| dstanek | morganfainberg: bknudson: are we really getting all of the users from the backend and then iterating over them to filter? | 20:10 |
| *** mdomsch has joined #openstack-dev | 20:10 | |
| morganfainberg | but that is a much further reaching argument. and i am aiming for the "fix things one item at a time" instead of "here is a massive change, deal with it" | 20:10 |
| morganfainberg | dstanek, yes. | 20:10 |
| *** heckj has joined #openstack-dev | 20:10 | |
| morganfainberg | dstanek, which is a bad assumption that you can do that i hope to correct in this dev cycle | 20:10 |
| *** neoXsys has joined #openstack-dev | 20:10 | |
| bknudson | we should require domain to be disabled before delete it | 20:10 |
| *** heckj has quit IRC | 20:10 | |
| morganfainberg | bknudson, already is the case | 20:10 |
| bknudson | dstanek: user_refs = self.identity_api.list_users() | 20:11 |
| morganfainberg | bknudson, can't delete a domain that is not disabled afaict | 20:11 |
| *** heckj has joined #openstack-dev | 20:11 | |
| bknudson | morganfainberg: great! | 20:11 |
| *** vkozhukalov has quit IRC | 20:11 | |
| bknudson | dstanek: actually does list_users multiple times... | 20:11 |
| dstanek | bknudson: yes :-( | 20:11 |
| morganfainberg | i shall be "fixing" that in more changes down the line. | 20:12 |
| morganfainberg | i hope | 20:12 |
| bknudson | and it's not just the users for the domain either, it's all users in all domains. | 20:12 |
| morganfainberg | bknudson, sort-of. | 20:13 |
| morganfainberg | well for the "primary" idp | 20:13 |
| morganfainberg | would miss all secondary idps. | 20:13 |
| dstanek | bknudson: right, that has got to hurt on big installations | 20:13 |
| morganfainberg | bknudson, but the per-domain-identity stuff is only partially working as is | 20:14 |
| *** harlowja_away is now known as harlowja | 20:14 | |
| morganfainberg | bknudson, http://git.openstack.org/cgit/openstack/keystone/tree/keystone/identity/controllers.py#n640 enabled domains can't be deleted | 20:14 |
| bknudson | maybe what's missing is a "service" layer between the controllers and managers | 20:14 |
| morganfainberg | bknudson, perhaps. but the controllers already _mostly_ provide that. | 20:15 |
| morganfainberg | they have the dependancies etc. | 20:15 |
| dstanek | bknudson: isn't that what the manager layer is responsible for? | 20:15 |
| *** denis_makogon_ has joined #openstack-dev | 20:16 | |
| morganfainberg | dstanek, it's split between controllers and managers | 20:16 |
| *** sgordon has joined #openstack-dev | 20:16 | |
| morganfainberg | basically, i think we should make a determination that it's either a manager purview or a controller purview | 20:16 |
| dstanek | morganfainberg: imo the controllers shouldn't have any real logic beyond web related stuff | 20:16 |
| dsantos_ | hi, I'm trying to get users by project using keystone v3. I saw in the api-ref that I should do a GET to v3/projects/​{project_id}​/users but I got a 404. Could someone help me? | 20:16 |
| *** dolphm has joined #openstack-dev | 20:17 | |
| *** Drankis has quit IRC | 20:17 | |
| bknudson | dstanek: right, but look at https://review.openstack.org/#/c/62197/5/keystone/identity/controllers.py _delete_domain_contents | 20:17 |
| bknudson | it's talking to identity and tokens for example | 20:17 |
| *** alexpilotti has joined #openstack-dev | 20:17 | |
| morganfainberg | i am not convinced we need an extra layer to do cross-manager talking. | 20:18 |
| bknudson | the controllers are doing way too much | 20:18 |
| *** heckj has left #openstack-dev | 20:18 | |
| morganfainberg | bknudson, dstanek , its a tough situation because of the direction things are headed with federation etc | 20:19 |
| *** vipul-away is now known as vipul | 20:19 | |
| bknudson | I agree with dstanek -- assuming he thinks the controllers should just be taking the request and forwarding it to a service layer | 20:19 |
| bknudson | service layer can talk to the managers | 20:19 |
| *** aelkikhia has joined #openstack-dev | 20:19 | |
| *** markmc has joined #openstack-dev | 20:19 | |
| *** denis_makogon_ is now known as denis_makogon | 20:19 | |
| *** aelkikhia has quit IRC | 20:19 | |
| morganfainberg | bknudson, so should @protect still occur at the controller? | 20:19 |
| dstanek | bknudson: yes, that's it; i was saying earlier that we may want to push some of that logic into the managers instead of duplicating it in the controller | 20:20 |
| *** mtaylor is now known as mordred | 20:20 | |
| morganfainberg | so what we're talking about here is making what "most" of the controllers are now a service layer, and adding a "request handler" above it | 20:20 |
| morganfainberg | ? | 20:21 |
| morganfainberg | and rename current "controlelrs" to service layer and new request handler the new controller layer? | 20:21 |
| *** dolphm has quit IRC | 20:21 | |
| * morganfainberg thinks this wont be doable in Icehouse. | 20:21 | |
| bknudson | I just don't think that the managers are the right place for it since they're independent. | 20:22 |
| bknudson | morganfainberg: yes, that's essentially it. | 20:22 |
| bknudson | I see it more as pushing what controllers are doing now into a service layer. | 20:22 |
| morganfainberg | strictly from a volume of change perspective | 20:22 |
| morganfainberg | hrm. actually | 20:22 |
| dstanek | in smaller steps i think pushing controller logic into the manager (which right now i see as that services layer) | 20:22 |
| bknudson | it's a refactoring so would give us something to do for m3. | 20:22 |
| *** Drankis has joined #openstack-dev | 20:22 | |
| morganfainberg | dstanek, i don't want to push logic down to the managers just to pull it back up | 20:23 |
| morganfainberg | dstanek, that is "make work" | 20:23 |
| dstanek | morganfainberg: where would you pull it up to? | 20:23 |
| bknudson | for some reason our v3 controllers are subclasses of v2 controllers now, aren't they? | 20:23 |
| bknudson | so they can share code. | 20:24 |
| morganfainberg | bknudson, yes =/ | 20:24 |
| dstanek | bknudson: yes | 20:24 |
| morganfainberg | ok, i think i can do this in a quick smallish refactor | 20:24 |
| *** ayoung_xc-ski is now known as ayoung | 20:24 | |
| morganfainberg | that just sets the stage for shuffling the code up/down where appropriate | 20:24 |
| dstanek | morganfainberg: cool | 20:25 |
| *** johnthetubaguy has joined #openstack-dev | 20:25 | |
| ayoung | morganfainberg, sorry I'm late to the party. I think that Managers will become the real business logic, not the controllers. Controllers are for talkoing to the outside world, but internal call go manager to manager, not controller to controller | 20:25 |
| *** jkyle has joined #openstack-dev | 20:25 | |
| ayoung | which, as I read up, is about where you guys landed | 20:26 |
| morganfainberg | ayoung, i am convinced we need another "service layer" | 20:26 |
| ayoung | nope | 20:26 |
| ayoung | manager is biz logic | 20:26 |
| morganfainberg | ayoung, managers are pivot points for drivers | 20:26 |
| ayoung | controller is binding to the outside world | 20:26 |
| ayoung | but not just | 20:26 |
| jkyle | is there a call sequence diagram for nova volume-attach operations? Like how it's passed around, who consumes what from whom, and at which stage? | 20:26 |
| morganfainberg | ayoung, i think we're lacking a clear definition here though | 20:27 |
| *** eglynn has joined #openstack-dev | 20:27 | |
| morganfainberg | ayoung, there are cases where the manager "layer" is insufficient | 20:27 |
| ayoung | morganfainberg, let the complexity emerge...yes, there might be the occasional time it is called for, we will see. | 20:27 |
| morganfainberg | ayoung, notably the whole "identity" aspect where identity is losing a lot of the reliability with the federation movement | 20:27 |
| ayoung | morganfainberg, you can always have a manager call not reflected in the underlying driver | 20:28 |
| morganfainberg | reliability of data being avaialble directly | 20:28 |
| morganfainberg | fair enough, i think we're going to see the opposite problem we're seeing now as we move there. | 20:28 |
| *** epim has joined #openstack-dev | 20:31 | |
| *** vipul is now known as vipul-away | 20:31 | |
| bknudson | ayoung: we already have operations too complicated for manager - see https://review.openstack.org/#/c/62197/5/keystone/identity/controllers.py | 20:32 |
| bknudson | _delete_domain_contents | 20:32 |
| bknudson | it operates on multiple managers -- so which manager would have the logic? | 20:32 |
| *** rfolco has quit IRC | 20:33 | |
| *** johnthetubaguy has quit IRC | 20:33 | |
| *** jecarey has joined #openstack-dev | 20:33 | |
| *** mrunge has joined #openstack-dev | 20:33 | |
| *** SumitNaiksatam has quit IRC | 20:34 | |
| bknudson | I see the "service" layer as sitting under controllers and above managers... it would have operations like "domain_service.delete()" that the controllers would call rather than "assignment_api.delete_domain()" | 20:34 |
| ayoung | bknudson, I doubt it. | 20:35 |
| *** markwash has joined #openstack-dev | 20:35 | |
| morganfainberg | ayoung, so would _delete_domain_contents just go on the assignment manager? | 20:35 |
| ayoung | morganfainberg, yes, and we get users out of domains | 20:35 |
| morganfainberg | ayoung, working on it. | 20:35 |
| ayoung | morganfainberg, you were the one that realized it was a mistake to put them there. | 20:36 |
| morganfainberg | ayoung, trying to get the cleanup done so it's easy to remove them. | 20:36 |
| *** richardboswell has joined #openstack-dev | 20:36 | |
| ayoung | domains are top level assignement groupings... | 20:36 |
| morganfainberg | ayoung, which is where this patchset and the subsequent ones are headed | 20:36 |
| ayoung | so assignments should own them | 20:36 |
| ayoung | yep | 20:36 |
| morganfainberg | i can push this logic down in this cleanup patchset or i can do it later on. both will accomplish the same thing. i'm trying to keep the changesets as digestable as possible | 20:37 |
| ayoung | So really, the problem is that we have a tangled up model right now, not that we have something that the current manager abstraction can't solve. | 20:37 |
| ayoung | morganfainberg, you are doing things right AFAICT | 20:37 |
| * morganfainberg has a ton of work to do this cycle. | 20:38 | |
| *** eglynn has quit IRC | 20:38 | |
| bknudson | this is why I think it's too early to promise stability of manager interfaces. | 20:39 |
| *** gyee has joined #openstack-dev | 20:39 | |
| morganfainberg | bknudson, thankfully the public interfaces of the managers isn't too bad to maintain most of the issue(s) we're seeing are internal-workings. | 20:40 |
| morganfainberg | bknudson, i am also not oposed to raising an exception saying "this moved, no you can't has it anymore" for a cycle. | 20:40 |
| morganfainberg | if something can't be maintained. | 20:40 |
| morganfainberg | it's not about 100% maintaining interfaces, it's also about communicating what changed to out-of-tree developers | 20:41 |
| bknudson | I don't think we should mix the responsibility of "talking to different backends" (sql vs kvs) with "business logic" | 20:42 |
| *** epim has quit IRC | 20:42 | |
| *** vipul-away is now known as vipul | 20:43 | |
| morganfainberg | bknudson, the biggest argument i could make for a service layer is it would allow caching to really be much clearer at the pivot point vs somewhere in the middle of business logic. | 20:43 |
| *** pixelb has quit IRC | 20:44 | |
| morganfainberg | and it would make cache invalidations easier (biggest argument for the -complexity-of-code- standpoint) | 20:44 |
| bknudson | morganfainberg: the pivot point for caching being the managers? | 20:44 |
| morganfainberg | bknudson, yes. | 20:44 |
| morganfainberg | bknudson, put cache/invalidation logic in at that point. | 20:45 |
| morganfainberg | right now it's a mix | 20:45 |
| morganfainberg | same as the rest of the manager code. | 20:45 |
| *** nelsnelson has quit IRC | 20:45 | |
| morganfainberg | unless we are also looking to cache the results of business logic (which we ... usually don't) | 20:45 |
| *** nelsnelson has joined #openstack-dev | 20:46 | |
| bknudson | having a sane and obvious place to put caching logic is a good reason for a refactoring | 20:46 |
| dstanek | morganfainberg: if you split the business logic out of the managers would the managers end up just being pass thrus to the backend? | 20:46 |
| morganfainberg | dstanek, pivot + caching | 20:46 |
| morganfainberg | dstanek, if i were to do that split | 20:46 |
| morganfainberg | dstanek, if it didn't have caching logic, it would just pass through | 20:46 |
| morganfainberg | bknudson, that would also make it much easier to avoid the kwargs issue | 20:47 |
| dstanek | morganfainberg: interesting...i don't know enough about the current arch to have an opinion yet, but i do like what you are saying | 20:47 |
| *** sarob_ has joined #openstack-dev | 20:47 | |
| morganfainberg | no kwargs on the managers, means caching is always possible service layer might need optional args | 20:48 |
| morganfainberg | but it would be a lot less call <method> that calls _<method> just to avoid kwargs | 20:48 |
| morganfainberg | because memoization and invalidations with kwargs becomes a nightmare | 20:48 |
| *** epim has joined #openstack-dev | 20:49 | |
| *** drewlander has quit IRC | 20:49 | |
| bknudson | I think we're always going to have kwargs as long as we support "extra" fields? | 20:49 |
| morganfainberg | i _think_ it would be doable without a lot of making merges a nightmare. | 20:49 |
| *** sarob has quit IRC | 20:49 | |
| bknudson | (or the equivalent of kwargs) | 20:50 |
| morganfainberg | bknudson, and the service layer makes that easy to handle. since it does that work and passes fixed areguments to the backend | 20:50 |
| morganfainberg | bknudson, s/makes/wouldmake | 20:50 |
| *** itooon has quit IRC | 20:50 | |
| morganfainberg | bknudson, it's just about placing the optional args in a place that it wont affect the in-line caching | 20:51 |
| morganfainberg | bknudson, but this can still all be technically solved w/o the need of a service layer for now | 20:52 |
| morganfainberg | so. | 20:52 |
| morganfainberg | back to the original discussion | 20:52 |
| *** vipul is now known as vipul-away | 20:52 | |
| *** sarob_ has quit IRC | 20:52 | |
| morganfainberg | handling the refactors and direction | 20:52 |
| ayoung | since dolph is so set against hacking the v2.0 off the urls...we need a new endpoint for Keystone | 20:52 |
| morganfainberg | ayoung, i vote bbq_identity | 20:53 |
| *** itooon has joined #openstack-dev | 20:53 | |
| ayoung | jquery? http://benalman.com/projects/jquery-bbq-plugin/ | 20:53 |
| morganfainberg | bknudson, dstanek, how much refactoring now vs say post m2 | 20:53 |
| morganfainberg | i'm trying to clean things up that affect config and the like now... for obvious reasons | 20:54 |
| *** eglynn has joined #openstack-dev | 20:54 | |
| morganfainberg | ayoung, not what i was going for but sure | 20:54 |
| morganfainberg | and do i push that logic down to the assignment manager so i can move on with the cleanup needed to get users out of domains | 20:55 |
| morganfainberg | or is it fine as is for this patchset and more logic shuffle will occur in subsequent ones | 20:56 |
| *** dolphm has joined #openstack-dev | 20:56 | |
| *** jtomasek has quit IRC | 20:57 | |
| bknudson | morganfainberg: as long as your cleaning things up then we'll be getting closer to where we want to be. | 20:58 |
| ayoung | bknudson, ++ | 20:58 |
| morganfainberg | bknudson, and there is lots of cleanup planned. | 20:58 |
| bknudson | like removing creating controller WTFs. | 20:58 |
| *** Drankis has quit IRC | 20:59 | |
| morganfainberg | unfortunately, there is some user-facing terms we can't "get rid of" for a while | 20:59 |
| *** jtomasek has joined #openstack-dev | 20:59 | |
| morganfainberg | but that will come with time. | 20:59 |
| morganfainberg | bknudson, ayoung, dstanek, ok i'm going to ask we leave that review as is, and i'll continue with cleanups and get that logic pushed down into the manager down the line. | 21:00 |
| morganfainberg | i apologize for the slight duplication of code, it is only temporary though | 21:00 |
| morganfainberg | and we can discuss splitting pivot/caaching down away from biz logic when we get towards m3 | 21:01 |
| morganfainberg | sicne i think that change will make landing things like federation much harder | 21:01 |
| dstanek | morganfainberg: that works for me...i'll change my vote | 21:01 |
| *** epim has quit IRC | 21:02 | |
| morganfainberg | dstanek, bknudson, ayoung , btw, good convo here on this topic. feel like we're making progress. | 21:02 |
| *** jistr has quit IRC | 21:02 | |
| *** markvoelker1 has joined #openstack-dev | 21:02 | |
| dstanek | bknudson: thoughts on my comment on https://review.openstack.org/#/c/58311/ ? | 21:03 |
| ayoung | dolphm, if you nix the URL parsing, we need new endpoints | 21:03 |
| ayoung | specificallyt, we need ones that will allow discoverability to work, while leaving the old broken ones in place | 21:03 |
| *** d0ugal has quit IRC | 21:03 | |
| *** d0ugal has joined #openstack-dev | 21:04 | |
| *** radsy has joined #openstack-dev | 21:04 | |
| fabiog | morganfainberg, just pushed a new patch with your required changes, please have a look | 21:04 |
| *** etoews1 has joined #openstack-dev | 21:04 | |
| *** d0ugal has quit IRC | 21:04 | |
| *** d0ugal has joined #openstack-dev | 21:04 | |
| *** jasdeepH has joined #openstack-dev | 21:04 | |
| dstanek | ayoung: do you have a link to the gist (or whatever) we were discussing yesterday in regards to the API versioning | 21:04 |
| ayoung | https://blueprints.launchpad.net/python-keystoneclient/+spec/endpoint-versioning dstanek | 21:05 |
| *** mdomsch has quit IRC | 21:05 | |
| bknudson | dstanek: I mentioned in patch set 2 that I didn't think this was the right fix - https://review.openstack.org/#/c/58311/2/keystone/tests/fixtures/cache.py | 21:05 |
| *** jasdeepH has quit IRC | 21:05 | |
| bknudson | it's fixing the symptom and not the problem. | 21:05 |
| *** samuelbercovici has joined #openstack-dev | 21:05 | |
| etoews1 | dtroyer_zz: yt? | 21:06 |
| *** sandywalsh_ has quit IRC | 21:06 | |
| *** samuelbercovici has quit IRC | 21:06 | |
| *** topol has quit IRC | 21:06 | |
| bknudson | ayoung: the identity spec says what the version strings are, so while parsing the url is not ideal, it should work. | 21:06 |
| ayoung | bknudson, its a hack, I admit it. But its like a hack you use to get through a dense jungle vine...just don't hack your own foot | 21:07 |
| *** pberis has quit IRC | 21:07 | |
| dstanek | bknudson: did your load_backends patch ever merge? | 21:07 |
| *** fabiog has quit IRC | 21:07 | |
| *** dims has quit IRC | 21:08 | |
| bknudson | dstanek: do any of my patches ever merge?? | 21:08 |
| dstanek | bknudson: that code isn't about fixing the symtom at all it's about moving setup/teardown into a fixture | 21:09 |
| *** dims has joined #openstack-dev | 21:09 | |
| dstanek | bknudson: i think you are right about the change that needs to happen, but that's another patch for another day | 21:09 |
| *** glenng has quit IRC | 21:10 | |
| *** yamahata has joined #openstack-dev | 21:10 | |
| dstanek | bknudson: i'm going to see if i can use that as a way to setup the cache; it provides a great seam for initialization code | 21:10 |
| *** eglynn has quit IRC | 21:10 | |
| bknudson | dstanek: but it also moves "del cache.REGION.backend" from teardown to setUp | 21:10 |
| bknudson | dstanek: https://review.openstack.org/#/c/58311/4/keystone/tests/core.py does "del cache.REGION.backend" in tearDown | 21:11 |
| *** jimjiang has quit IRC | 21:11 | |
| dstanek | bknudson: i can introduce a patch that moves it first, but it wouldn't change the content of my patch | 21:11 |
| bknudson | dstanek: actually, the load_backends change did merge. | 21:11 |
| dstanek | bknudson: fixtures will typically put teardown things in the setup; that's the pattern on how they work | 21:12 |
| bknudson | dstanek: the patch that moves it would be fixing the symptom and not the problem. | 21:12 |
| *** jimjiang has joined #openstack-dev | 21:12 | |
| *** vipul-away is now known as vipul | 21:12 | |
| dstanek | bknudson: https://pypi.python.org/pypi/fixtures#creating-fixtures | 21:13 |
| dstanek | bknudson: i'm taking the exact same logic that already exists and moving it into a fixture | 21:13 |
| bknudson | dstanek: don't you do addCleanup to do cleanup stuff? | 21:14 |
| ayoung | bknudson, it is extra stupid for us, cuz keystoneclient already knows the correct auth_url to use to get the token. We don';t even need the service catalog | 21:15 |
| bknudson | there's an example on that page: self.addCleanup(delattr, self, 'frobnozzle') | 21:15 |
| *** sgordon has quit IRC | 21:16 | |
| dstanek | bknudson: ah i see...if i remove the two lines that make the tests work would that change your mind? | 21:16 |
| bknudson | dstanek: yes | 21:17 |
| *** pberis has joined #openstack-dev | 21:17 | |
| bknudson | dstanek: well, you'd remove the two lines and replace it with self.addCleanup() | 21:17 |
| *** sandywalsh_ has joined #openstack-dev | 21:17 | |
| *** jimjiang has quit IRC | 21:17 | |
| dstanek | bknudson: so that mean that the tests would still be broken | 21:18 |
| *** jimjiang has joined #openstack-dev | 21:18 | |
| bknudson | dstanek: right, this change is a refactoring and not a fix | 21:19 |
| dstanek | bknudson: part of setting up the test is making sure that the cache is in the correct state to begin the test - that's all that does | 21:19 |
| dstanek | bknudson: i'll break it up into 2 commits | 21:20 |
| *** nati_ueno has quit IRC | 21:20 | |
| *** d0ugal has quit IRC | 21:20 | |
| bknudson | dstanek: how was the cache in the correct state before? | 21:20 |
| dstanek | bknudson: before my patch? | 21:21 |
| bknudson | dstanek: yes, as the code is now... the cache isn't in the correct state so how do the tests pass? | 21:21 |
| *** tdruiva_ has joined #openstack-dev | 21:22 | |
| dstanek | bknudson: they pass on accident when run as a part of the full test suite; they fail individually; https://bugs.launchpad.net/keystone/+bug/1251300 | 21:22 |
| uvirtbot | Launchpad bug 1251300 in keystone "test_cache_layer_*_crud depend on test execution order" [Low,In progress] | 21:22 |
| *** jimjiang has quit IRC | 21:23 | |
| dstanek | a happy accident, but still and accident | 21:23 |
| morganfainberg | bknudson, dstanek , the first test can't use caching basically | 21:23 |
| morganfainberg | dstanek's fix solves that | 21:23 |
| *** e0ne has quit IRC | 21:23 | |
| *** jimjiang has joined #openstack-dev | 21:23 | |
| morganfainberg | it's a silly race i didn't account for in the testing suite | 21:23 |
| *** urulama has joined #openstack-dev | 21:24 | |
| *** d0ugal has joined #openstack-dev | 21:24 | |
| *** d0ugal has joined #openstack-dev | 21:24 | |
| morganfainberg | race-like condition | 21:24 |
| *** d0ugal has quit IRC | 21:24 | |
| morganfainberg | or uhm sometrhing | 21:24 |
| *** yolanda has quit IRC | 21:24 | |
| dstanek | something in the test suite configures the cache before those tests run | 21:24 |
| bknudson | I think the problem is that we've got "stuff" happening at import time. | 21:24 |
| dstanek | when you run them in isolation that's not happening | 21:24 |
| bknudson | and we shouldn't be doing that "stuff" at import time. | 21:24 |
| *** tdruiva has quit IRC | 21:25 | |
| dstanek | bknudson: i agree and i think your load_backends seam would be a great place to fix it | 21:25 |
| morganfainberg | dstanek, it's in services.py that does it | 21:25 |
| morganfainberg | dstanek, i think | 21:25 |
| dstanek | bknudson: also we are talking about global state because of how the decorator works | 21:25 |
| *** xqueralt has quit IRC | 21:25 | |
| morganfainberg | dstanek, and the tests don't load the config until afterwards unlike the real-keystone startup | 21:25 |
| *** yassine has joined #openstack-dev | 21:26 | |
| bknudson | global state :( | 21:26 |
| morganfainberg | bknudson, sadly, hard to get around that in some cases. | 21:26 |
| dstanek | morganfainberg: i believe you are right about services.py | 21:27 |
| morganfainberg | bknudson, notably since caching memoization decorator is an instance-bound method to the cache region | 21:27 |
| bknudson | dstanek: well, if that's what we have to do... at least take a look at a change in load_backends. | 21:27 |
| morganfainberg | dstanek, +awesome to any cleanup making that better :) | 21:27 |
| dstanek | i feel like today has been a productive day and i haven | 21:28 |
| *** nati_ueno has joined #openstack-dev | 21:28 | |
| dstanek | 't written much code yet | 21:28 |
| *** urulama has quit IRC | 21:28 | |
| morganfainberg | dstanek, this is whayt the summit felt like to me. | 21:28 |
| *** jimjiang has quit IRC | 21:29 | |
| morganfainberg | dsantos_, i saw a question you were asking higher up. did you figure it out? | 21:29 |
| morganfainberg | dsantos_, sorry just realized i hadn't responded to you yet. | 21:29 |
| *** sgordon has joined #openstack-dev | 21:29 | |
| *** sgordon has quit IRC | 21:29 | |
| *** sgordon has joined #openstack-dev | 21:29 | |
| *** d0ugal has joined #openstack-dev | 21:30 | |
| *** d0ugal has joined #openstack-dev | 21:30 | |
| *** d0ugal has quit IRC | 21:30 | |
| *** jimjiang has joined #openstack-dev | 21:30 | |
| *** colinmcnamara has joined #openstack-dev | 21:30 | |
| *** d0ugal has joined #openstack-dev | 21:31 | |
| *** d0ugal has joined #openstack-dev | 21:31 | |
| *** d0ugal has quit IRC | 21:31 | |
| dstanek | bknudson: looks like it merged two days ago...i'll start to play | 21:31 |
| ayoung | dolphm, OK, so what if we put all of the available versions on both the /v2.0 sand /v3 pages? THat way, even if I give the wrong auth_url, we can run discovery and find the right one | 21:32 |
| *** razumovalex has joined #openstack-dev | 21:32 | |
| *** gszasz has quit IRC | 21:32 | |
| bknudson | ayoung: I like that idea. | 21:32 |
| bknudson | (and I also don't have a problem with parsing url) | 21:33 |
| *** d0ugal has joined #openstack-dev | 21:33 | |
| ayoung | bknudson, it would have to be something like "other_versions" | 21:33 |
| ayoung | which is a new API change and all that | 21:33 |
| ayoung | could be in the links, though | 21:33 |
| *** glenng has joined #openstack-dev | 21:35 | |
| *** jimjiang has quit IRC | 21:35 | |
| *** d0ugal has joined #openstack-dev | 21:35 | |
| *** d0ugal has joined #openstack-dev | 21:35 | |
| *** d0ugal has quit IRC | 21:35 | |
| *** jimjiang has joined #openstack-dev | 21:36 | |
| bknudson | ayoung: wondering if we need it in the /v3 response? | 21:37 |
| bknudson | I thought the problem was just existing catalogs that have /v2.0? | 21:37 |
| ayoung | bknudson, true | 21:37 |
| *** mikeoutland has joined #openstack-dev | 21:38 | |
| ayoung | bknudson, what we really need is to be able to look at the response and say "this is not the version I want" | 21:38 |
| *** d0ugal has joined #openstack-dev | 21:38 | |
| *** d0ugal has joined #openstack-dev | 21:38 | |
| *** d0ugal has quit IRC | 21:38 | |
| *** dprince has quit IRC | 21:38 | |
| bknudson | ayoung: so you'll request /v2.0 ... how does it know that it's v2 or v3? and how does it know what version you wanted? | 21:39 |
| *** briancline has quit IRC | 21:39 | |
| ayoung | bknudson, cuz you asked for a v3 client | 21:39 |
| *** d0ugal has joined #openstack-dev | 21:39 | |
| *** d0ugal has quit IRC | 21:39 | |
| ayoung | bknudson, this is so flipping broken....all of discovery is broken, and we let ourselves get hungf up on this? | 21:40 |
| ayoung | ARGH! | 21:40 |
| ayoung | Its not like I had an approach that would fix this or anthing either. | 21:40 |
| *** glenng1 has joined #openstack-dev | 21:40 | |
| ayoung | grumble | 21:40 |
| *** d0ugal has joined #openstack-dev | 21:41 | |
| *** d0ugal has joined #openstack-dev | 21:41 | |
| *** d0ugal has quit IRC | 21:41 | |
| bknudson | ayoung: the v2 response does say "id": "v2.0" | 21:41 |
| bknudson | ayoung: but I thought you wanted to avoid a round-trip to the server? | 21:41 |
| bknudson | putting the v3 endpoint in the v2.0 version response would help with discovery | 21:41 |
| *** d0ugal has joined #openstack-dev | 21:41 | |
| *** d0ugal has quit IRC | 21:41 | |
| *** d0ugal has joined #openstack-dev | 21:41 | |
| ayoung | bknudson, I jsut want to hack the URLs. | 21:42 |
| ayoung | I don';t want to playu oin this horse and poiny show | 21:42 |
| * ayoung can't type too annoyed | 21:42 | |
| *** glenng has quit IRC | 21:42 | |
| *** dvarga has quit IRC | 21:42 | |
| ayoung | here is the thing...we say "h, you have to do discovery" but then all of the rest of theAPI is based on a-priori knowledge | 21:43 |
| *** mikeoutland has quit IRC | 21:43 | |
| *** mriedem has joined #openstack-dev | 21:43 | |
| *** razumovalex has quit IRC | 21:44 | |
| bknudson | we only have version discovery and not all API discovery | 21:45 |
| *** neelashah has quit IRC | 21:45 | |
| termie_ | ayoung: here here brother | 21:46 |
| *** termie_ is now known as termie | 21:46 | |
| *** termie has quit IRC | 21:47 | |
| *** termie has joined #openstack-dev | 21:47 | |
| * ayoung summoned a termie, | 21:47 | |
| *** d0ugal has joined #openstack-dev | 21:48 | |
| *** d0ugal has joined #openstack-dev | 21:48 | |
| *** d0ugal has quit IRC | 21:48 | |
| *** d0ugal has joined #openstack-dev | 21:48 | |
| *** d0ugal has joined #openstack-dev | 21:49 | |
| *** d0ugal has quit IRC | 21:49 | |
| *** d0ugal has joined #openstack-dev | 21:49 | |
| *** yfried has joined #openstack-dev | 21:49 | |
| termie | ayoung: be careful what you wish for | 21:49 |
| dstanek | ayoung: are there examples of other APIs that do discovery? | 21:50 |
| ayoung | termie, you mean I'll end up with OCL? | 21:50 |
| termie | ayoung: you'll end up with that anyway | 21:50 |
| *** dave_tucker has quit IRC | 21:50 | |
| ayoung | dstanek, APIs don' | 21:50 |
| ayoung | t do sicovery, clients do | 21:50 |
| termie | ayoung: i just haven't read all the scrollback so i may not actually be agreeing with you :p | 21:50 |
| ayoung | APIs don't do discovery, clients do | 21:50 |
| *** glenng1 is now known as glenng | 21:50 | |
| yfried | asking again here (redirected from infra): | 21:50 |
| yfried | Hi guys, I'm looking to get the external connectivity (ie - ping 8.8.8.8) on the devstack-neutron gate so this patch can work: https://review.openstack.org/#/c/55146/ | 21:50 |
| yfried | can any one help me? | 21:50 |
| yfried | or to put it another way - are we sure the gate doesn't have external connection at the moment? | 21:50 |
| *** mikeoutland has joined #openstack-dev | 21:51 | |
| dstanek | ayoung: right, i meant that support (or encourage) | 21:51 |
| *** jasondotstar has quit IRC | 21:51 | |
| ayoung | termie, heh, the problem is that we have two versions of the API, v2.0 and v3, and we've been telling everyone fo so long that they need to register then endpoints with v2.0, and now we want to tell them go to v3, and they are all "No way." | 21:51 |
| *** briancline has joined #openstack-dev | 21:52 | |
| termie | v3 = vLulz | 21:52 |
| ayoung | dstanek, Yeah, its called a web brower. We can't actually use one with Keystone because we hand back broken links. Which I tried to fix way back when | 21:52 |
| ayoung | termie, I knew you were going to say that | 21:52 |
| ayoung | cuz you said it before | 21:52 |
| *** BLZbubba has joined #openstack-dev | 21:52 | |
| termie | ayoung: repitition is the key to propaganda | 21:52 |
| ayoung | termie, I am thinking that version numbers in URLs are not the right way to go | 21:52 |
| *** Darkness650 has joined #openstack-dev | 21:52 | |
| *** MaxV has joined #openstack-dev | 21:53 | |
| Alexei_987 | dhellmann: Hi do you have some time to chat? | 21:53 |
| *** bknudson has quit IRC | 21:53 | |
| *** rnirmal has quit IRC | 21:53 | |
| dhellmann | Alexei_987: I'm in the release project meeting at the moment | 21:53 |
| dstanek | ayoung: your definition of discovery seems to be different than conneg - at least in my current understanding | 21:53 |
| termie | ayoung: my gut assumption is that we could have dealt with versions in a better, backwards compatible way | 21:53 |
| * dhellmann has too many meetings :-/ | 21:53 | |
| *** vartom111119 has quit IRC | 21:53 | |
| Alexei_987 | dhellmann: when you'll be available? | 21:53 |
| dhellmann | Alexei_987: 10 minutes or so? | 21:54 |
| ayoung | termie, I've seen some recommendations that it should be handled via content_types. Something bothers me about that, can't quite put my finger on it, but it seems better than what we do now | 21:54 |
| Alexei_987 | dhellmann: cool, ping me when ready | 21:54 |
| *** bknudson has joined #openstack-dev | 21:54 | |
| ayoung | But something in the request should say "I want the old version blah" but the URL is the good name of the object | 21:54 |
| ayoung | shouldn't be the url | 21:55 |
| *** ohnoimdead has joined #openstack-dev | 21:56 | |
| termie | hmm, i think i wrote an update to this with explicit descriptions of how to do versioning: https://wiki.openstack.org/wiki/EasyApi | 21:56 |
| *** yeylon__ has quit IRC | 21:56 | |
| termie | anyway, that's cool, you guys go do your own thing | 21:56 |
| termie | i'll go back to making it as irrelevant as possible by coding around it in ocl ;) | 21:57 |
| Apsu | The more standards the better! | 21:57 |
| Apsu | Open as in choice, yo | 21:57 |
| dstanek | ayoung: it would have to be a combination of content-types and a hypertext driven workflow | 21:57 |
| termie | oo, that update is allllmost 3 years old | 21:57 |
| termie | vishy: oo, almost the 3-year anniversary of EasyApi | 21:58 |
| *** e0ne has joined #openstack-dev | 21:58 | |
| termie | vishy: drinks on the 28th? ;) | 21:58 |
| *** jcoufal has joined #openstack-dev | 21:59 | |
| ayoung | termie, while I hope you are successful, Einstein said it so well "There is a race between mankind and the universe. Mankind is trying to build bigger, better, faster, and more foolproof machines. The universe is trying to build bigger, better, and faster fools. So far the universe is winning." | 22:00 |
| *** e0ne has quit IRC | 22:00 | |
| *** nati_uen_ has joined #openstack-dev | 22:01 | |
| termie | ayoung: "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." | 22:01 |
| ayoung | Sounds like your driving philosophy | 22:01 |
| termie | just some random einstein quote that is probably made up | 22:02 |
| dhellmann | Alexei_987: ready, if you are | 22:02 |
| termie | but seems to point in a more productive direction than yours | 22:02 |
| Alexei_987 | dhellmann: I'm here | 22:02 |
| *** nati_uen_ has quit IRC | 22:03 | |
| Alexei_987 | dhellmann: working on one patch I've became bored of passing context object all around | 22:03 |
| *** SumitNaiksatam has joined #openstack-dev | 22:03 | |
| Alexei_987 | dhellmann: what do you think about following approach - https://review.openstack.org/#/c/62639/2/ironic/common/context.py | 22:03 |
| *** branen has quit IRC | 22:03 | |
| *** nati_ueno has quit IRC | 22:03 | |
| *** nati_uen_ has joined #openstack-dev | 22:03 | |
| dhellmann | Alexei_987: have a look at the locals module in oslo-incubator | 22:03 |
| *** dstanek has quit IRC | 22:04 | |
| Alexei_987 | dhellmann: yep :) | 22:04 |
| dhellmann | Alexei_987: also consider whether this new "global" makes testing more difficult | 22:04 |
| Alexei_987 | why? | 22:04 |
| *** melwitt has quit IRC | 22:04 | |
| *** melwitt has joined #openstack-dev | 22:04 | |
| dhellmann | Alexei_987: as a general rule, I would prefer an explicit argument | 22:04 |
| *** mfer has quit IRC | 22:04 | |
| termie | Alexei_987: because most of testing revolves around overriding conditions in specific cases | 22:05 |
| Alexei_987 | dhellmann: but it's passed through 10 methods and never used | 22:05 |
| ayoung | termie, mayhaps...I didn't design keystone, just came along later and tried to do with it what needed to be done. I wouldn't claim that I was the right person to do this, just that there was a vacuum, and a need. I never wanted to do all this anyway. I wanted to be... | 22:05 |
| ayoung | a lumberjack | 22:05 |
| termie | ayoung: have you seen my beard? | 22:05 |
| *** colinmcnamara has quit IRC | 22:05 | |
| termie | ayoung: http://instagram.com/p/iCZIJIKdiH/ | 22:05 |
| dhellmann | Alexei_987: it introduces something else to mock out (or clear) rather than inject from a test method | 22:05 |
| ayoung | termie, no. I think I've only seen you with the mustache | 22:06 |
| Alexei_987 | dhellmann: https://review.openstack.org/#/c/62639/2/ironic/tests/conductor/test_manager.py | 22:06 |
| termie | (i just got a haircut) | 22:06 |
| *** melwitt1 has joined #openstack-dev | 22:06 | |
| dhellmann | Alexei_987: it sounds like maybe the stack is overly complex, then? | 22:06 |
| *** alunduil has quit IRC | 22:06 | |
| *** melwitt has quit IRC | 22:06 | |
| Alexei_987 | dhellmann: well in ironic it makes sense, cause most of the stuff is executed with used context, and some stuff requires admin context | 22:06 |
| Alexei_987 | dhellmann: so it can escalate privileges for 1 action and release just after | 22:07 |
| dhellmann | Alexei_987: sure, I get why it's needed, I'm just saying 10 levels that don't use it seems like a lot of layers of abstraction | 22:08 |
| morganfainberg | termie, oh hi! | 22:08 |
| termie | morganfainberg: good day to you, sir | 22:08 |
| ayoung | termie, Ok, so the problem is the service catalog which we return in the Keystone token. Would you recommend we drop that? I know you suggested Zookeeper or a compatable service registration/discovery point,m byut I never quite got your larger vision | 22:08 |
| Alexei_987 | dhellmann: well we have a lot of abstractions around | 22:08 |
| dhellmann | Alexei_987: yep :-) | 22:08 |
| *** ArxCruz has joined #openstack-dev | 22:09 | |
| termie | ayoung: which part of it is the problem? | 22:09 |
| termie | ayoung: i can tell you my vision, but it'd be useful to get the succinct backstory | 22:09 |
| Alexei_987 | dhellmann: so what do you think in general? does pros outweigh the cons? | 22:09 |
| dhellmann | Alexei_987: I still prefer an explicit argument :-/ | 22:09 |
| *** melwitt1 has quit IRC | 22:10 | |
| Alexei_987 | dhellmann: Ok yet another patch abandoned | 22:10 |
| *** melwitt has joined #openstack-dev | 22:10 | |
| ayoung | termie, all of the operations that the keystone client performs are based on the URLs that come back from the endpoints list. We've been telling people to put v2.0 in them | 22:10 |
| *** markmcclain has quit IRC | 22:10 | |
| Alexei_987 | dhellmann: thanks clearing this out | 22:10 |
| dhellmann | Alexei_987: sorry | 22:10 |
| *** jdob has quit IRC | 22:11 | |
| Alexei_987 | dhellmann: no problem it was just a draft proposal | 22:11 |
| ayoung | and since people have been doing this, and writing clients against Keystone that work on this assumption, if we change the endpoints (drop the versions) we break all the clients out there | 22:11 |
| dhellmann | Alexei_987: ok | 22:11 |
| *** Darkness650 has left #openstack-dev | 22:11 | |
| termie | ayoung: the catalog is returning v2, or the clients are using v2 instead of what the catalog returns? | 22:11 |
| ayoung | the catalog returns http://host:500/v2.0 | 22:12 |
| termie | ayoung: and if it returned http://host:5000/v3.11 what would happen? | 22:12 |
| *** DennyZhang has joined #openstack-dev | 22:13 | |
| ayoung | well, a v3.11 client would work fine, and everyone else would be broken | 22:13 |
| termie | ayoung: or http://host:2001? | 22:13 |
| termie | ayoung: why would the client be broken? you just mean that the client assumes the version is 2? | 22:13 |
| *** pberis has quit IRC | 22:13 | |
| ayoung | so, that is what we would want, and that, right now, breaks everything, but we can fix that...we have the discovery mechanism to find out what versions are supported, but it is not used by default yet | 22:13 |
| termie | ayoung: if it responds correctly to their request there shouldn't be an issue, right? | 22:13 |
| *** pberis has joined #openstack-dev | 22:14 | |
| *** davidhadas_ has quit IRC | 22:14 | |
| termie | ayoung: so, right now there is a v2 and a v3 way to request a token, i assume | 22:14 |
| ayoung | yes | 22:14 |
| ayoung | but requesting a token is done a-priori | 22:14 |
| ayoung | you set the AUTH_URL | 22:14 |
| ayoung | and call a specific client | 22:14 |
| *** athomas has quit IRC | 22:14 | |
| *** pberis has quit IRC | 22:14 | |
| termie | ayoung: and then we discover other endpoints, some of which may have versions you know about and some of which may not | 22:15 |
| ayoung | right...I think all of the other services are doing versioned endpoints due to our bad example | 22:15 |
| *** bswartz has quit IRC | 22:15 | |
| termie | ayoung: don't worry, it isn't keystone itself's bad example | 22:15 |
| termie | ayoung: just rackspace's | 22:16 |
| termie | ayoung: but regardless of the versioning bit, the answer is either a) smarter responses or b) better clients | 22:16 |
| ayoung | so I want to start with b | 22:16 |
| morganfainberg | termie, a bit of both a and b? | 22:16 |
| ayoung | IE...look at the url, and if it has a version on it, hack it off | 22:16 |
| *** d0ugal has quit IRC | 22:17 | |
| termie | morganfainberg: sometimes, but then you can run into things like html vs browsers | 22:17 |
| *** xmltok_ has joined #openstack-dev | 22:17 | |
| termie | morganfainberg: and quirksmode | 22:17 |
| termie | where nobody knows the statemachine that will be traversed to decide the answer | 22:17 |
| *** joesavak has quit IRC | 22:17 | |
| ayoung | we can make the server smarter over time, but client needs to work with older servers, and needs to be smart today | 22:17 |
| *** pcm__ has joined #openstack-dev | 22:17 | |
| *** ruhe has quit IRC | 22:17 | |
| termie | so with b, you have stuff that will do it's damndest to figure out the correct call and have extensive tests written to make sure it works against everything, like Universal RSS Parser in the day | 22:17 |
| *** sgordon has quit IRC | 22:18 | |
| morganfainberg | termie, fair point. but i'd like somewhat smarter responses... it's personal view. | 22:18 |
| morganfainberg | termie, oh good example | 22:18 |
| *** markmc has quit IRC | 22:18 | |
| termie | OCL, btw, has probably a good handle on that because it was written while refusing to work on the APIs | 22:18 |
| termie | because i've washed my hands of them | 22:18 |
| *** d0ugal has joined #openstack-dev | 22:18 | |
| *** d0ugal has joined #openstack-dev | 22:18 | |
| *** d0ugal has quit IRC | 22:18 | |
| termie | on the smarter response side we can try to guess which client is asking, sort of like a User-Agent in the web | 22:19 |
| termie | i don't really like the way websites do that, because it tends to fuck up when i am writing test stuff | 22:19 |
| *** teran has joined #openstack-dev | 22:19 | |
| *** d0ugal has joined #openstack-dev | 22:19 | |
| *** d0ugal has quit IRC | 22:19 | |
| *** d0ugal has joined #openstack-dev | 22:19 | |
| termie | and it menas tons of my code has inserted false user-agents over the years | 22:19 |
| *** dave_tucker has joined #openstack-dev | 22:19 | |
| morganfainberg | termie, i like the concept of "supports capability of" being sent in | 22:20 |
| termie | so, i've tended to lean towards b | 22:20 |
| *** d0ugal has joined #openstack-dev | 22:20 | |
| *** d0ugal has joined #openstack-dev | 22:20 | |
| *** d0ugal has quit IRC | 22:20 | |
| morganfainberg | and the server giving something useful back based upon that | 22:20 |
| morganfainberg | defaulting to "basic" functionality | 22:20 |
| termie | sure, additional info lets client decide, but older stuff is still an issue | 22:20 |
| termie | in farmboy, the thing i've been working on recently, i still have a config value for whether you are supporting neutron or not | 22:20 |
| morganfainberg | termie, right. | 22:20 |
| *** d0ugal has joined #openstack-dev | 22:20 | |
| *** d0ugal has quit IRC | 22:20 | |
| *** d0ugal has joined #openstack-dev | 22:20 | |
| *** pcm has quit IRC | 22:20 | |
| *** d0ugal has joined #openstack-dev | 22:21 | |
| *** d0ugal has quit IRC | 22:21 | |
| *** d0ugal has joined #openstack-dev | 22:21 | |
| morganfainberg | termie, it's a question of where to stick the logic | 22:21 |
| termie | morganfainberg: yup, i think i made that clear in my previous statements ;) | 22:21 |
| *** d0ugal has joined #openstack-dev | 22:21 | |
| morganfainberg | termie, i like setting a minimum bar (sorry Ie7, go away) and expand from there | 22:21 |
| morganfainberg | offering optional/better handling (Responses) if supported by the client, and let the client decide | 22:21 |
| termie | morganfainberg: somebody will still end up writing universal rss parser and being well loved for it | 22:22 |
| morganfainberg | again, totally personal bias | 22:22 |
| termie | morganfainberg: or at least tolerated at your conferences | 22:22 |
| morganfainberg | termie, i know. ;) | 22:22 |
| termie | i apparently am being called into a coffee run | 22:22 |
| morganfainberg | termie, anyway i'll let you get back to the original topic now that we've tangented some | 22:22 |
| *** d0ugal has joined #openstack-dev | 22:22 | |
| *** d0ugal has joined #openstack-dev | 22:23 | |
| *** d0ugal has joined #openstack-dev | 22:23 | |
| *** d0ugal has quit IRC | 22:23 | |
| termie | ayoung: brb after coffee and will explain the zookeeper vision | 22:23 |
| *** d0ugal has joined #openstack-dev | 22:23 | |
| *** d0ugal has quit IRC | 22:23 | |
| *** d0ugal has joined #openstack-dev | 22:23 | |
| ayoung | termie, cool | 22:23 |
| *** d0ugal has joined #openstack-dev | 22:24 | |
| *** d0ugal has quit IRC | 22:24 | |
| *** d0ugal has joined #openstack-dev | 22:24 | |
| *** jobewan has quit IRC | 22:25 | |
| *** d0ugal has joined #openstack-dev | 22:25 | |
| *** d0ugal has quit IRC | 22:25 | |
| *** d0ugal has joined #openstack-dev | 22:25 | |
| *** pcm__ has quit IRC | 22:26 | |
| *** glenng has quit IRC | 22:26 | |
| *** yassine has quit IRC | 22:28 | |
| *** d0ugal has joined #openstack-dev | 22:28 | |
| *** d0ugal has quit IRC | 22:28 | |
| *** d0ugal has joined #openstack-dev | 22:28 | |
| *** jimjiang has quit IRC | 22:28 | |
| *** jimjiang has joined #openstack-dev | 22:29 | |
| *** boris-42 has quit IRC | 22:29 | |
| *** d0ugal has joined #openstack-dev | 22:30 | |
| *** d0ugal has joined #openstack-dev | 22:30 | |
| *** d0ugal has quit IRC | 22:30 | |
| *** d0ugal has joined #openstack-dev | 22:33 | |
| *** nermina has quit IRC | 22:33 | |
| *** d0ugal has joined #openstack-dev | 22:34 | |
| *** d0ugal has quit IRC | 22:34 | |
| *** d0ugal has joined #openstack-dev | 22:34 | |
| *** d0ugal has joined #openstack-dev | 22:35 | |
| *** d0ugal has joined #openstack-dev | 22:35 | |
| *** d0ugal has quit IRC | 22:35 | |
| *** nermina has joined #openstack-dev | 22:38 | |
| *** d0ugal has joined #openstack-dev | 22:38 | |
| *** d0ugal has joined #openstack-dev | 22:38 | |
| *** d0ugal has quit IRC | 22:38 | |
| *** d0ugal has joined #openstack-dev | 22:40 | |
| *** danielbruno has quit IRC | 22:41 | |
| *** aeperezt has quit IRC | 22:42 | |
| *** jmontemayor has joined #openstack-dev | 22:44 | |
| *** sarob has joined #openstack-dev | 22:44 | |
| *** DennyZhang has quit IRC | 22:44 | |
| *** gordc has quit IRC | 22:45 | |
| *** SergeyLukjanov has quit IRC | 22:46 | |
| *** stevemar has quit IRC | 22:50 | |
| *** troytoman is now known as troytoman-away | 22:50 | |
| termie | ayoung: I RETURN | 22:50 |
| *** jmontemayor has quit IRC | 22:51 | |
| termie | ayoung: the idea with the zookeeper etc thing is that the real solution we are trying to solve here is a place to store reliable information about where to locate the things in our system | 22:51 |
| *** dave_tucker has quit IRC | 22:51 | |
| termie | ayoung: and there already exist significantly more reliable / available codebases for doing just that then some random crap we're going to add to keystone | 22:52 |
| *** dave_tucker has joined #openstack-dev | 22:52 | |
| morganfainberg | termie, ++ | 22:52 |
| termie | ayoung: that already deal in things like paxos and fault tolerance | 22:52 |
| *** thomasem has quit IRC | 22:52 | |
| termie | ayoung: so why invent a broken wheel | 22:52 |
| termie | ayoung: zookeeper was based on models in use by large companies already, whereas i haven't heard anybody on keystone besides maybe myself pushing for any of those ideas | 22:53 |
| termie | ayoung: if somebody is running a large cloud, they need a reliable service in the middle to find this information | 22:53 |
| Apsu | If only we had a lightweight protocol for accessing a sort of index. A directory perhaps. | 22:54 |
| termie | ayoung: and additionally they tend to need a way to do things like outtages and drain out a datacenter | 22:54 |
| *** loq_mac has joined #openstack-dev | 22:55 | |
| *** buzztroll has quit IRC | 22:56 | |
| *** yfried has quit IRC | 22:56 | |
| *** sarob has quit IRC | 22:57 | |
| *** aeperezt has joined #openstack-dev | 22:58 | |
| termie | ayoung: so, long story shorter, have a scheme for storing service information in zookeeper, teach openstack services how to use that, probably let keystone shim some information in front of it, and eventually define the "catalog" service in terms of it | 22:58 |
| *** dave_tucker has quit IRC | 22:59 | |
| *** johnthetubaguy has joined #openstack-dev | 22:59 | |
| *** dave_tucker has joined #openstack-dev | 22:59 | |
| *** jimjiang has quit IRC | 23:00 | |
| *** dims has quit IRC | 23:01 | |
| *** jimjiang has joined #openstack-dev | 23:01 | |
| *** jcoufal has quit IRC | 23:01 | |
| *** MaxV has quit IRC | 23:01 | |
| *** jpich has quit IRC | 23:01 | |
| *** denis_makogon has quit IRC | 23:02 | |
| *** arnaud__ has quit IRC | 23:02 | |
| *** arnaud has quit IRC | 23:02 | |
| *** harlowja is now known as harlowja_away | 23:02 | |
| *** mrunge has quit IRC | 23:02 | |
| ayoung | I return from shoveling...and read up | 23:02 |
| *** nkinder has quit IRC | 23:03 | |
| ayoung | termie, what does zookeeper have going for it that LDAP does not? | 23:03 |
| *** iccha_ is now known as iccha-dinner | 23:03 | |
| *** pablosan has quit IRC | 23:04 | |
| *** kevinconway has quit IRC | 23:04 | |
| *** MaxV has joined #openstack-dev | 23:04 | |
| *** johnthetubaguy has quit IRC | 23:04 | |
| ayoung | I know that "regions" is where we first discussed this. I am guessing the zookeeper has a smart way of saying "here is the right service for someone in your location" | 23:04 |
| *** melwitt1 has joined #openstack-dev | 23:05 | |
| *** melwitt has quit IRC | 23:05 | |
| *** kbringard has quit IRC | 23:05 | |
| *** mikeoutland has quit IRC | 23:05 | |
| *** harlowja_away is now known as harlowja | 23:05 | |
| *** MaxV has quit IRC | 23:05 | |
| *** Farooque has joined #openstack-dev | 23:05 | |
| *** mikeoutland has joined #openstack-dev | 23:06 | |
| termie | ayoung: i barely know how to answer that question, do any of these overlap with ldap? http://en.wikipedia.org/wiki/Apache_ZooKeeper#Typical_use_cases | 23:06 |
| *** lsmola_ has quit IRC | 23:06 | |
| ayoung | termie, yeah. the D in LDAP is Directory. LDAP is just the protocol. | 23:06 |
| ayoung | You stick things in LDAP that you want to find later | 23:06 |
| *** jtomasek has quit IRC | 23:07 | |
| ayoung | users, printers, Queus | 23:07 |
| *** eharney has quit IRC | 23:07 | |
| ayoung | Queues | 23:07 |
| termie | ayoung: i didn't see the word directory on that page | 23:07 |
| morganfainberg | ayoung, zookeeper is more dynamic. | 23:07 |
| ayoung | morganfainberg, I get the sense that it is more workload focused | 23:07 |
| morganfainberg | ayoung, as in, i see LDAP as a more permanent store | 23:07 |
| ayoung | termie, naming service and directory are often used interchangably, which is why I brought it up | 23:08 |
| termie | ayoung: is it just that you've worked with ldap so much that you always talk about? almost nobody else ever does | 23:08 |
| Ryan_Lane | I don't see how zookeeper or ldap are alike at all | 23:08 |
| termie | ayoung: it has done a job in the past, one it is still used for, which is listing email addresses in outlook | 23:08 |
| ayoung | termie, its from the enterprise world...not the startup world | 23:08 |
| termie | Ryan_Lane: smaes | 23:08 |
| termie | Ryan_Lane: sames | 23:08 |
| Ryan_Lane | ldap is just another datastore | 23:09 |
| ayoung | I am not at all familiaer with Zookeeper, which is why I am asking | 23:09 |
| Ryan_Lane | it's a permanent datastore that has a very strict schema | 23:09 |
| termie | ayoung: well i feel like if you read that short wikipedia page you'll have a better idea | 23:09 |
| Ryan_Lane | and an awkward protocol :D | 23:09 |
| morganfainberg | termie, so am i understanding this correctly: basically register a service w/ ZK on startup (etc) and derive catalog/endpoints based upon that. if you start the service elsewhere it registers and is available | 23:10 |
| *** david-lyle_ has quit IRC | 23:10 | |
| morganfainberg | termie, rely on that as the "catalog" vs the templates or ... sql or whatever in ks now | 23:10 |
| *** dstanek has joined #openstack-dev | 23:10 | |
| morganfainberg | termie, or am i making an extra leap here that isn't needed | 23:10 |
| morganfainberg | ? | 23:10 |
| termie | morganfainberg: in reality it is usually used almost opposite, the thing starting the serivce looks to zk on startup to find out where it should go | 23:10 |
| morganfainberg | termie, ah | 23:10 |
| morganfainberg | termie, ok i was inverting it | 23:11 |
| *** burt has quit IRC | 23:11 | |
| ayoung | termie, I did...I read it back when you first told me about it. Here is the things that have come up recently that I think we need to sort out: we have huge number of machines running 'stack and there is no centralized naming or location services for them. Then, for the end uers, they have the smae problem, but * 1000 cuz they are using virtual machines | 23:11 |
| termie | morganfainberg: but it is also used in the way you describe | 23:11 |
| morganfainberg | termie, i can see it being useful in both ways within OS. | 23:11 |
| *** e0ne has joined #openstack-dev | 23:11 | |
| ayoung | so my question was: are you thinking zookeeper as a service that end users get from the openstack deployment, or something for internal use, or for both? | 23:11 |
| termie | morganfainberg: yup | 23:11 |
| morganfainberg | ayoung, i would (at least initially) see it as internal and it gets used to generate the data provided to end-users | 23:12 |
| termie | ayoung: something they get in an openstack deployment that they'll probably want to base internal things off of as well | 23:12 |
| termie | ayoung: i think initially we'll still be shimming the data into keystone | 23:12 |
| termie | because ocean boiling | 23:12 |
| *** vijendar has quit IRC | 23:12 | |
| *** mikeoutland has quit IRC | 23:13 | |
| *** vijendar has joined #openstack-dev | 23:13 | |
| termie | but the protocol can easily be used directly | 23:13 |
| glucas | termie: you mean zookeeper initially as a layer on top of keystone? | 23:13 |
| termie | glucas: nay, i just mean things still use the catalog | 23:13 |
| glucas | ah | 23:13 |
| shardy | ayoung: https://wiki.openstack.org/wiki/Heat/Blueprints/InstanceUsers | 23:13 |
| termie | glucas: so initially just start building that catalog from the zk | 23:13 |
| ayoung | termie, so would it make sense that each nova compute node would be registered with zookeeper? Cells? | 23:14 |
| morganfainberg | glucas, that is exactly where i saw it. | 23:14 |
| *** nermina has quit IRC | 23:14 | |
| shardy | ayoung: If you can add anything, pleas do, particularly to clarify the x509 section ;) | 23:14 |
| morganfainberg | glucas, ^ | 23:14 |
| termie | ayoung: maybe? it definitely makes sense that the service locations are | 23:14 |
| ayoung | shardy, listen in to termie 's conversation... | 23:14 |
| *** cpallares has quit IRC | 23:14 | |
| morganfainberg | termie, ayoung, we could use auth_token middleware to register the service...it's dirty hack but.... | 23:15 |
| *** thomasem has joined #openstack-dev | 23:15 | |
| termie | ayoung: i think once people learn how the tool works they will be able to figure out how much data is reasonable ot put in there | 23:15 |
| morganfainberg | and derive keystone's location from that connection to zk as well | 23:15 |
| termie | morganfainberg: just have the ervices register themselves? | 23:15 |
| morganfainberg | termie, possibly | 23:15 |
| ayoung | morganfainberg, question of trust | 23:15 |
| *** e0ne has quit IRC | 23:16 | |
| *** dims has joined #openstack-dev | 23:16 | |
| morganfainberg | ayoung, it's not fully baked, might have that issue. depends on the control ZK can provide | 23:16 |
| termie | registering of services is something that is done on the order of 1 request every minute or hour or day | 23:16 |
| *** thomasem has quit IRC | 23:16 | |
| termie | that's not a hard scalability goal to hit | 23:16 |
| *** thomasem has joined #openstack-dev | 23:16 | |
| ayoung | morganfainberg, its the same general problem as freeipa: you want to know a new X is coming in, so you pre-allocate a resource for it, but you also give X a secret, one time use, that it can use to register | 23:17 |
| morganfainberg | termie, ayoung, i need to spend some more time looking into ZK to know how feasable it would be, but it might be possible. | 23:18 |
| termie | morganfainberg: how feasible what is? | 23:18 |
| morganfainberg | termie, auto-registration could be. | 23:18 |
| ayoung | termie, so I think I've been addressing the same problem, but using older technologies. LDAP and FreeIPA do a lot of the same things. Gives you Kerberos based SSO and X509 management. I'm trying to see where a FreeIPA based approach would fall down, and what the scalability benefits of Zookeeper are. | 23:18 |
| morganfainberg | termie, or at least from the ephemeral 'child' sense. | 23:19 |
| morganfainberg | in zk | 23:19 |
| termie | morganfainberg: it is as easy as an authenticated call to your keystone service registry thing at the moment, no? | 23:19 |
| ayoung | LDAP assumes a fairly static set up, but should be able to handle the physical side...I was thinking that the virtual machine side might be too dynamic | 23:19 |
| morganfainberg | termie, probably | 23:19 |
| *** yamahata has quit IRC | 23:20 | |
| termie | move the backend, then move the frontend == backwards compat | 23:20 |
| *** vijendar has quit IRC | 23:20 | |
| ekarlso | strptime('17/Dec/2013:20:11:51 +0100', '%d/%b/%Y:%X %z') < works in py3 but 2.7 ? | 23:20 |
| *** _dbalog has quit IRC | 23:20 | |
| *** sgordon has joined #openstack-dev | 23:21 | |
| *** alunduil has joined #openstack-dev | 23:21 | |
| ayoung | termie, I was also unclean because Zookeeper calls itself a configuration service, and I was wondering where the line was between what you do with Zookeeper and what you do with Puppet. | 23:21 |
| termie | ayoung: not particularly related | 23:22 |
| *** reed has quit IRC | 23:22 | |
| sudorandom | +1 for service registry for OpenStack in general. Zookeeper is likely not the only viable option though so I'd keep that in mind. I read 'zookeeper' during this conversation as a number of possible tools. Here's one possible alternative: https://github.com/coreos/etcd | 23:23 |
| termie | ayoung: configuration as in it stores the configuration you look up, for example if the things starting services look at it at startup to know who to talk to | 23:23 |
| *** thomasem has quit IRC | 23:23 | |
| termie | ayoung: where as puppet is sort of an integrated configration application (as in it applies configurations) service | 23:23 |
| ekarlso | termie: can I bother you boss to ask what the discussion is? like a 30k foot view | 23:24 |
| *** urulama has joined #openstack-dev | 23:24 | |
| ayoung | ekarlso, read up..started with me asking about his vision for integrating with zookeeper | 23:25 |
| ayoung | ekarlso, which came up in the "regions" context | 23:25 |
| termie | ekarlso: we're discussing how things like the service catalog should be stored/updated/interpreted | 23:25 |
| termie | ekarlso, ayoung: came up in the regions context because i made the claim that the regions solution is solving a symptom not a cause | 23:26 |
| *** xmltok_ has quit IRC | 23:26 | |
| *** shardy is now known as shardy_afk | 23:26 | |
| ayoung | termie, so, instead of an auth_url, we would give the client a Zookeeper url. The client would ask "where is keystone" and get a token from there, but no service catalog. THen it would go to zookeeper and say "now where is nova"? | 23:26 |
| termie | ayoung: probably not, no | 23:26 |
| ekarlso | ah ok | 23:27 |
| termie | ayoung: you would give an auth_url that is a keystone instance, depending on your internal architecture that is probably routed based on config loaded from zk in your internal bits, to a load balanced keystone, that thing tosses you a token as normal with a catalog that was generated from zk | 23:28 |
| ayoung | termie, so Keystone would front zookeeper to the end user | 23:28 |
| termie | ayoung: if we are removing the catalog lookup from the token call, you would have a separate call to a catlaog service that builds the catalog from zk | 23:28 |
| ayoung | zookeeper would be an internal service that the end use does not talk to directly, at least not for Open Stack related calls. | 23:28 |
| morganfainberg | termie, which i think is a good approach, remove the catalog from the token that is | 23:29 |
| termie | ayoung: yeah, end user is not going to talk to it | 23:29 |
| *** urulama has quit IRC | 23:29 | |
| termie | ayoung: internal users mights | 23:29 |
| termie | ayoung: but that is probably out of scope | 23:29 |
| ayoung | morganfainberg, I always saw that as a performance tweak | 23:29 |
| termie | it is a useful tool so people tend to use it for things once they find out it is there | 23:29 |
| *** asalkeld has quit IRC | 23:29 | |
| *** radsy has quit IRC | 23:29 | |
| *** asalkeld has joined #openstack-dev | 23:30 | |
| sudorandom | I think of service registry as an internal service similar to where rabbitmq fits in. | 23:30 |
| ayoung | termie, I was thinking about it in terms of end users. It looks like it is used for long running tasks, Hadooplike | 23:30 |
| termie | morganfainberg: i think the catalog is fine in the token, personally, because plenty of things will need the data nayway, i don't reall ylike the format but it seaves a common call | 23:30 |
| *** xmltok_ has joined #openstack-dev | 23:30 | |
| termie | ayoung: it is used for things like, "my service is starting up and wants to know where the memcache and db backends it can connect to are" | 23:31 |
| termie | ayoung: which is basically internal discovery | 23:31 |
| *** buzztroll has joined #openstack-dev | 23:31 | |
| *** xmltok_ has quit IRC | 23:31 | |
| ayoung | termie, OK, you answered the big question. It is an internal service. I agree, we need one. | 23:32 |
| morganfainberg | termie, the issue is that it actually has gotten too large in some cases and causes errors with header sizes | 23:32 |
| * ayoung needs to go be a dad now. | 23:32 | |
| *** xmltok_ has joined #openstack-dev | 23:32 | |
| morganfainberg | termie, we've had to construct it as "optional" in the token data in some cases | 23:32 |
| termie | morganfainberg: well, that is probably solvable by adding things asking for specific endpoints to the token call | 23:32 |
| morganfainberg | termie, fair point, perhaps it would be nice ot have the option to reference a stand-alone service catalog | 23:33 |
| morganfainberg | termie, again, client / use case specific | 23:33 |
| termie | morganfainberg: aye, i think a standalone service is not a bad idea | 23:33 |
| morganfainberg | *shrug* things to think about when we have a central service. | 23:33 |
| termie | morganfainberg: are you stuffing the entire response into a header? | 23:33 |
| morganfainberg | termie, PKI tokens kindof do that | 23:33 |
| termie | morganfainberg: part of the point is that keystone should not be a single instance | 23:33 |
| morganfainberg | termie, i agree with that | 23:34 |
| morganfainberg | termie, 100% | 23:34 |
| *** jmontemayor has joined #openstack-dev | 23:34 | |
| termie | things like ZK can actually be central services because they are designed to be | 23:34 |
| termie | but keystone should spawn a bunch of instances that get appropriately routed by your network config | 23:35 |
| morganfainberg | termie, totally makes sense to me. | 23:35 |
| termie | which, in various places, are actually also configured by zk | 23:35 |
| termie | (because it is a popular place to store that kind of data) | 23:36 |
| termie | sudorandom: re other options, definitely look at them and come up with some pros and cons | 23:37 |
| termie | sudorandom: zookeeper, for all its warts, has tended to be the one teams i've been part of have come to decide on | 23:37 |
| termie | sudorandom: so that's my initial guess at the right solution | 23:38 |
| *** p4tux has quit IRC | 23:39 | |
| *** p4tux has joined #openstack-dev | 23:39 | |
| *** BLZbubba has quit IRC | 23:39 | |
| *** glucas has quit IRC | 23:39 | |
| *** glucas has joined #openstack-dev | 23:40 | |
| *** romcheg has quit IRC | 23:41 | |
| *** mikeoutland has joined #openstack-dev | 23:42 | |
| *** sthaha has joined #openstack-dev | 23:42 | |
| *** tdruiva_ has quit IRC | 23:44 | |
| *** tdruiva has joined #openstack-dev | 23:44 | |
| jamielennox | damn, missed all of that - i like the idea of a central service catalog that isn't keystone and pushing that out of the token, it's important then i think that we can fetch an entire global service catalog and expect the services to extract the right one from the list | 23:44 |
| jamielennox | but that is something that is fairly easily re-used | 23:45 |
| *** branen has joined #openstack-dev | 23:46 | |
| *** rcleere has quit IRC | 23:47 | |
| *** dolphm has quit IRC | 23:47 | |
| *** jmontemayor has quit IRC | 23:48 | |
| *** tdruiva has quit IRC | 23:49 | |
| *** dstanek has quit IRC | 23:51 | |
| *** jhesketh__ has joined #openstack-dev | 23:53 | |
| *** fbo is now known as fbo_away | 23:55 | |
| *** SumitNaiksatam has quit IRC | 23:56 | |
| *** faramir1 has joined #openstack-dev | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!