| *** hjensas has joined #openstack-ironic | 00:01 | |
| *** aoxen has joined #openstack-ironic | 00:12 | |
| openstackgerrit | ShangXiao proposed openstack/ironic master: [api-ref] Add fault names for the error response codes https://review.openstack.org/538717 | 00:27 |
|---|---|---|
| *** hshiina has joined #openstack-ironic | 00:28 | |
| openstackgerrit | fupingxie proposed openstack/ironic-lib master: Add attemps for the partprobe command https://review.openstack.org/542035 | 00:55 |
| *** liuzz has joined #openstack-ironic | 00:55 | |
| *** phuongnh has joined #openstack-ironic | 00:56 | |
| *** r-daneel has joined #openstack-ironic | 01:06 | |
| *** tiendc has joined #openstack-ironic | 01:14 | |
| *** r-daneel_ has joined #openstack-ironic | 01:32 | |
| *** r-daneel has quit IRC | 01:33 | |
| *** r-daneel_ is now known as r-daneel | 01:33 | |
| *** yufei has joined #openstack-ironic | 01:45 | |
| *** pmannidi has joined #openstack-ironic | 02:33 | |
| openstackgerrit | Jinxing Fang proposed openstack/sushy master: Add system status field https://review.openstack.org/550669 | 02:46 |
| openstackgerrit | Jinxing Fang proposed openstack/sushy master: Add system status field https://review.openstack.org/550669 | 03:03 |
| *** hjensas has quit IRC | 03:08 | |
| *** hjensas has joined #openstack-ironic | 03:08 | |
| *** rbudden has joined #openstack-ironic | 03:20 | |
| *** namnh has joined #openstack-ironic | 03:21 | |
| *** rcernin_ has joined #openstack-ironic | 03:23 | |
| *** rcernin has quit IRC | 03:25 | |
| openstackgerrit | wangqi proposed openstack/ironic master: Duplicate explicit target name: "statsd" https://review.openstack.org/554095 | 03:25 |
| *** rcernin_ has quit IRC | 03:29 | |
| *** rcernin has joined #openstack-ironic | 03:29 | |
| *** rcernin has quit IRC | 03:33 | |
| *** zhangfei has joined #openstack-ironic | 03:37 | |
| *** pmannidi has quit IRC | 03:38 | |
| *** pmannidi has joined #openstack-ironic | 03:42 | |
| *** rcernin has joined #openstack-ironic | 03:49 | |
| *** zhangfei has quit IRC | 03:55 | |
| openstackgerrit | Jinxing Fang proposed openstack/sushy master: Add system status field https://review.openstack.org/550669 | 03:55 |
| *** Nisha_Agarwal has joined #openstack-ironic | 04:17 | |
| *** rwsu has joined #openstack-ironic | 04:24 | |
| *** rwsu has quit IRC | 04:25 | |
| openstackgerrit | zenghui.shi proposed openstack/ironic master: BIOS Settings: Add DB model https://review.openstack.org/511162 | 04:31 |
| openstackgerrit | zenghui.shi proposed openstack/ironic master: BIOS Settings: Add DB API https://review.openstack.org/511402 | 04:31 |
| openstackgerrit | zenghui.shi proposed openstack/ironic master: BIOS Settings: Add RPC API https://review.openstack.org/512200 | 04:31 |
| *** zshi_ has quit IRC | 04:33 | |
| *** zshi_ has joined #openstack-ironic | 04:33 | |
| *** zshi_ is now known as zshi | 04:34 | |
| openstackgerrit | Dao Cong Tien proposed openstack/ironic-tempest-plugin master: Rescue mode standalone tests https://review.openstack.org/538119 | 04:36 |
| *** itlinux has quit IRC | 04:39 | |
| openstackgerrit | Dao Cong Tien proposed openstack/ironic master: Update CI jobs for rescue mode https://review.openstack.org/528704 | 04:40 |
| *** dpawar has joined #openstack-ironic | 05:23 | |
| *** rbudden has quit IRC | 05:34 | |
| *** rbartal has joined #openstack-ironic | 05:46 | |
| *** openstackgerrit has quit IRC | 05:49 | |
| *** dpawar_ has joined #openstack-ironic | 06:02 | |
| *** dpawar has quit IRC | 06:02 | |
| *** marios has joined #openstack-ironic | 06:12 | |
| *** Nisha_Agarwal has quit IRC | 06:16 | |
| *** zhangfei has joined #openstack-ironic | 06:22 | |
| *** ianychoi__ has joined #openstack-ironic | 06:26 | |
| *** ianychoi_ has quit IRC | 06:29 | |
| *** zhangfei has quit IRC | 06:32 | |
| *** jaganathan has joined #openstack-ironic | 06:42 | |
| *** fragatina has quit IRC | 06:42 | |
| *** openstackgerrit has joined #openstack-ironic | 06:46 | |
| openstackgerrit | Tuan Luong-Anh proposed openstack/ironic master: [WIP] Implement iRMC BIOS configuration https://review.openstack.org/534595 | 06:46 |
| openstackgerrit | Dao Cong Tien proposed openstack/ironic master: Update CI jobs for rescue mode https://review.openstack.org/528704 | 06:49 |
| *** jcoufal has joined #openstack-ironic | 06:57 | |
| *** pmannidi has quit IRC | 06:58 | |
| *** d0ugal has joined #openstack-ironic | 07:00 | |
| *** d0ugal has quit IRC | 07:00 | |
| *** d0ugal has joined #openstack-ironic | 07:00 | |
| *** rcernin has quit IRC | 07:07 | |
| *** mjura has joined #openstack-ironic | 07:20 | |
| *** yolanda_ is now known as yolanda | 07:21 | |
| *** zhangfei has joined #openstack-ironic | 07:22 | |
| *** zhangfei has quit IRC | 07:26 | |
| *** pcaruana has joined #openstack-ironic | 07:39 | |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/networking-generic-switch master: WIP Use native Zuul job https://review.openstack.org/553813 | 07:47 |
| *** e0ne has joined #openstack-ironic | 07:52 | |
| *** AlexeyAbashkin has joined #openstack-ironic | 07:54 | |
| *** OctopusZhang_ has joined #openstack-ironic | 07:55 | |
| openstackgerrit | Nguyen Hai proposed openstack/ironic master: Follow the new PTI for document build https://review.openstack.org/553473 | 07:57 |
| *** yufei has quit IRC | 07:59 | |
| openstackgerrit | Nguyen Hai proposed openstack/python-ironicclient master: Follow the new PTI for document build https://review.openstack.org/554157 | 08:01 |
| *** dpawar_ has quit IRC | 08:09 | |
| *** zhangfei has joined #openstack-ironic | 08:16 | |
| *** tesseract has joined #openstack-ironic | 08:17 | |
| *** racedo has joined #openstack-ironic | 08:20 | |
| *** jaganathan has quit IRC | 08:25 | |
| *** jaganathan has joined #openstack-ironic | 08:26 | |
| *** zhangfei has quit IRC | 08:27 | |
| *** dpawar has joined #openstack-ironic | 08:34 | |
| openstackgerrit | Nguyen Hai proposed openstack/python-ironicclient master: Follow the new PTI for document build https://review.openstack.org/554157 | 08:36 |
| *** jaganathan_ has joined #openstack-ironic | 08:38 | |
| *** jaganathan has quit IRC | 08:38 | |
| *** jaganathan_ has quit IRC | 08:42 | |
| *** stendulker has joined #openstack-ironic | 08:45 | |
| *** jaganathan_ has joined #openstack-ironic | 08:54 | |
| *** jtomasek has joined #openstack-ironic | 09:02 | |
| *** athomas has joined #openstack-ironic | 09:11 | |
| *** hshiina has quit IRC | 09:13 | |
| *** MattMan has quit IRC | 09:15 | |
| *** MattMan has joined #openstack-ironic | 09:15 | |
| *** zhangfei has joined #openstack-ironic | 09:17 | |
| *** hoonetorg has quit IRC | 09:19 | |
| *** ajya has joined #openstack-ironic | 09:27 | |
| *** mgoddard has joined #openstack-ironic | 09:29 | |
| *** hoonetorg has joined #openstack-ironic | 09:32 | |
| *** OctopusZhang_ has quit IRC | 09:37 | |
| *** exodusftw has quit IRC | 09:37 | |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic master: Fix ``agent`` deploy interface to call ``boot.prepare_instance`` https://review.openstack.org/499050 | 09:38 |
| *** exodusftw has joined #openstack-ironic | 09:39 | |
| *** OctopusZhang_ has joined #openstack-ironic | 09:44 | |
| *** OctopusZhang_ has quit IRC | 09:48 | |
| openstackgerrit | Dao Cong Tien proposed openstack/ironic-tempest-plugin master: Rescue mode standalone tests https://review.openstack.org/538119 | 09:51 |
| *** namnh has quit IRC | 10:05 | |
| *** sambetts_ is now known as sambetts | 10:08 | |
| *** zhangfei has quit IRC | 10:09 | |
| skazi | hi guys | 10:11 |
| skazi | if anyone has a spare minute please take a look at https://review.openstack.org/#/c/553330/ | 10:11 |
| patchbot | patch 553330 - ironic (stable/pike) - Use internal Swift endpoint by default | 10:11 |
| skazi | especially I'd like to know your opinion about the conf option vs hardcoded 'internal' | 10:12 |
| *** dtantsur|afk is now known as dtantsur | 10:19 | |
| dtantsur | morning ironic | 10:19 |
| dtantsur | sambetts, stendulker, hi, please check https://review.openstack.org/553841 it fixes the inspector CI | 10:19 |
| patchbot | patch 553841 - ironic - Gate: stop setting IRONIC_ENABLED_INSPECT_INTEFACE... | 10:19 |
| *** e0ne has quit IRC | 10:19 | |
| dtantsur | in combination with https://review.openstack.org/553797 | 10:20 |
| patchbot | patch 553797 - ironic-inspector - devstack: enable "fake" inspection implementation | 10:20 |
| pas-ha | skazi: how will it work with internal? If this code is used when constructing the tempurl, I presume there generally should not be any routing from the provisioning network to the network holding internal API endpoints | 10:21 |
| skazi | pas-ha: currently it uses public endpoints in pike and internal,public in later releases | 10:22 |
| skazi | afaik in newton it was also internal | 10:23 |
| sambetts | pas-ha: actually generally in the deployments I do the provisioning network has access to the control plane but no route to the outside world or the public endpoints | 10:23 |
| skazi | note that the exact bug we hit was with iLO not being able to pull the vmedia images from public endpoints | 10:24 |
| skazi | so it's not really about provisioning network but the iLO network not having routing to public | 10:24 |
| skazi | it might overlap but doesn't have to I guess | 10:24 |
| *** e0ne has joined #openstack-ironic | 10:25 | |
| skazi | this is the only case I know where we need to worry about connecting from bmc network to openstack control plane | 10:25 |
| skazi | normally e.g. for power control you only need the other way around | 10:25 |
| *** jcoufal_ has joined #openstack-ironic | 10:32 | |
| *** phuongnh has quit IRC | 10:33 | |
| *** ajya has quit IRC | 10:33 | |
| *** zhangfei has joined #openstack-ironic | 10:33 | |
| *** jcoufal has quit IRC | 10:35 | |
| *** jcoufal has joined #openstack-ironic | 10:35 | |
| *** jcoufal_ has quit IRC | 10:38 | |
| *** tiendc has quit IRC | 10:39 | |
| *** jcoufal_ has joined #openstack-ironic | 10:40 | |
| *** jcoufal has quit IRC | 10:42 | |
| *** e0ne has quit IRC | 10:51 | |
| *** zhangfei has quit IRC | 10:51 | |
| *** stendulker has quit IRC | 11:17 | |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/networking-generic-switch master: WIP Use native Zuul job https://review.openstack.org/553813 | 11:22 |
| *** e0ne has joined #openstack-ironic | 11:23 | |
| *** e0ne has quit IRC | 11:40 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic stable/queens: Rework logic handling reserved orphaned nodes in the conductor https://review.openstack.org/554202 | 11:41 |
| *** e0ne has joined #openstack-ironic | 11:44 | |
| *** logan- has quit IRC | 11:52 | |
| *** logan- has joined #openstack-ironic | 11:52 | |
| *** bfournie has quit IRC | 11:53 | |
| *** bfournie has joined #openstack-ironic | 11:53 | |
| *** bfournie has quit IRC | 11:58 | |
| *** jlabarre has joined #openstack-ironic | 12:08 | |
| *** trown|outtypewww is now known as trown|ruck | 12:12 | |
| *** gyankum has joined #openstack-ironic | 12:22 | |
| *** dmellado has quit IRC | 12:24 | |
| *** bfournie has joined #openstack-ironic | 12:25 | |
| *** dmellado has joined #openstack-ironic | 12:27 | |
| openstackgerrit | Nguyen Hai proposed openstack/ironic master: Follow the new PTI for document build https://review.openstack.org/553473 | 12:35 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/networking-generic-switch master: WIP Use native Zuul job https://review.openstack.org/553813 | 12:36 |
| *** mjturek has joined #openstack-ironic | 12:45 | |
| openstackgerrit | Merged openstack/networking-generic-switch master: Remove config.get_config() https://review.openstack.org/553902 | 12:49 |
| *** rloo has joined #openstack-ironic | 12:55 | |
| *** dprince has joined #openstack-ironic | 12:57 | |
| *** baha has joined #openstack-ironic | 13:03 | |
| *** milan_ has joined #openstack-ironic | 13:03 | |
| *** dpawar has quit IRC | 13:11 | |
| *** cdearborn has joined #openstack-ironic | 13:15 | |
| *** cdearborn has quit IRC | 13:16 | |
| *** cdearborn has joined #openstack-ironic | 13:17 | |
| openstackgerrit | Merged openstack/ironic-ui stable/pike: Imported Translations from Zanata https://review.openstack.org/548780 | 13:22 |
| openstackgerrit | Merged openstack/ironic-ui stable/queens: Imported Translations from Zanata https://review.openstack.org/547896 | 13:22 |
| *** patchbot has quit IRC | 13:23 | |
| *** patchbot has joined #openstack-ironic | 13:23 | |
| *** Goneri has joined #openstack-ironic | 13:29 | |
| *** mjturek has quit IRC | 13:35 | |
| *** rbartal has quit IRC | 13:36 | |
| *** rbudden has joined #openstack-ironic | 13:37 | |
| *** jroll has quit IRC | 13:42 | |
| *** jroll has joined #openstack-ironic | 13:43 | |
| *** hamzy__ is now known as hamzy | 13:43 | |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/networking-generic-switch master: WIP Use native Zuul job https://review.openstack.org/553813 | 13:47 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/networking-generic-switch master: Rename tempest plugin folder https://review.openstack.org/554233 | 13:47 |
| *** rbartal has joined #openstack-ironic | 13:48 | |
| *** rbartal has quit IRC | 13:54 | |
| *** dtantsur is now known as dtantsur|brb | 13:56 | |
| jroll | morning y'all | 14:01 |
| *** beekneemech is now known as bnemec | 14:02 | |
| TheJulia | Good Morning everyone! | 14:02 |
| * TheJulia begins warming up the espresso machine | 14:02 | |
| rloo | Good monday morning TheJulia, jroll, pas-ha, dtantsur|brb | 14:11 |
| *** mjturek has joined #openstack-ironic | 14:11 | |
| * etingof wishes TheJulia to enjoy her espresso ;) | 14:12 | |
| *** rbartal has joined #openstack-ironic | 14:13 | |
| openstackgerrit | Merged openstack/ironic master: change python-libguestfs to python-guestfs for ubuntu https://review.openstack.org/551834 | 14:15 |
| openstackgerrit | Merged openstack/ironic-inspector master: Updated from global requirements https://review.openstack.org/553182 | 14:15 |
| openstackgerrit | Merged openstack/python-ironicclient master: Updated from global requirements https://review.openstack.org/553251 | 14:15 |
| *** namnh has joined #openstack-ironic | 14:20 | |
| *** mjturek has quit IRC | 14:21 | |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/networking-generic-switch master: Rename tempest plugin folder https://review.openstack.org/554233 | 14:26 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/networking-generic-switch master: WIP Use native Zuul job https://review.openstack.org/553813 | 14:26 |
| rloo | jroll: wrt the healthcheck endpoint -- we're not going to add a CLI for it, right? | 14:26 |
| jroll | rloo: correct | 14:26 |
| jroll | or rather, "please no" | 14:27 |
| rloo | jroll: i am going to update the rfe/bug with that. to record it :) | 14:27 |
| TheJulia | ++ to please no | 14:27 |
| rloo | :D | 14:27 |
| jroll | thanks | 14:27 |
| rloo | jroll: but would be good to update api-ref? | 14:27 |
| TheJulia | does the heathcheck plugin not have it's own api ref? | 14:28 |
| jroll | rloo: I don't think so, it's an optional middleware, the docs for it are available in oslo | 14:28 |
| jroll | https://docs.openstack.org/oslo.middleware/latest/reference/healthcheck_plugins.html | 14:28 |
| jroll | er | 14:28 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/networking-generic-switch master: WIP Use native Zuul job https://review.openstack.org/553813 | 14:28 |
| jroll | no, that's right | 14:28 |
| rloo | so we assume users of ironic can figure that out, that there is a healthcheck option? i think the only place is in code/release note. | 14:28 |
| rloo | config option i guess. | 14:29 |
| jroll | config, yeah | 14:29 |
| jroll | https://github.com/openstack/ironic/blob/master/ironic/conf/healthcheck.py#L23 | 14:30 |
| jroll | which points to docs :) | 14:30 |
| *** zshi has quit IRC | 14:30 | |
| *** mjturek has joined #openstack-ironic | 14:30 | |
| rloo | jroll: guess it is better than nothing. | 14:31 |
| *** crushil has joined #openstack-ironic | 14:31 | |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/networking-generic-switch master: Rename tempest plugin folder https://review.openstack.org/554233 | 14:35 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/networking-generic-switch master: WIP Use native Zuul job https://review.openstack.org/553813 | 14:35 |
| *** gyankum has quit IRC | 14:36 | |
| *** mjturek has quit IRC | 14:39 | |
| *** mjturek has joined #openstack-ironic | 14:42 | |
| rloo | we have ngs documentation being published now: https://docs.openstack.org/networking-generic-switch/latest/ | 14:43 |
| *** zshi has joined #openstack-ironic | 14:43 | |
| pas-ha | rloo: yay! :-) | 14:44 |
| TheJulia | \o/ | 14:44 |
| TheJulia | I was really surprised they were not being published last week, considering I remember goign to them at one point | 14:45 |
| rloo | TheJulia: yeah, me too. That's why I submitted the patch to do it :) | 14:45 |
| *** hjensas has quit IRC | 14:58 | |
| *** itlinux has joined #openstack-ironic | 15:00 | |
| radez | bfournie: hey could you help me work through double checking my aggregate setup? I'm getting this in my nova logs: 'AggregateInstanceExtraSpecsFilter: (start: 1, end: 0)' | 15:05 |
| radez | and a No host found error | 15:05 |
| *** dtantsur|brb is now known as dtantsur | 15:10 | |
| dtantsur | morning jroll, TheJulia, rloo | 15:11 |
| openstackgerrit | Nguyen Hai proposed openstack/python-ironicclient master: Follow the new PTI for document build https://review.openstack.org/554157 | 15:11 |
| jroll | \o dtantsur | 15:11 |
| rloo | dtantsur: :) | 15:12 |
| *** ElCoyote_ has joined #openstack-ironic | 15:15 | |
| *** zhangfei has joined #openstack-ironic | 15:15 | |
| dtantsur | jroll: opinion from pov of a bigger cloud needed: https://review.openstack.org/#/c/553308/1/specs/approved/power-fault-recovery.rst@52 | 15:20 |
| patchbot | patch 553308 - ironic-specs - Support power fault recovery | 15:20 |
| jroll | sure, will put it on my list | 15:20 |
| NobodyCam | Good Morning Ironic'ers | 15:46 |
| dtantsur | happy Monday NobodyCam | 15:47 |
| *** rbartal has quit IRC | 15:47 | |
| NobodyCam | Happy Monday dtantsur :) | 15:47 |
| NobodyCam | :p | 15:47 |
| *** zhangfei has quit IRC | 15:49 | |
| *** mjturek has quit IRC | 15:51 | |
| *** hw_wutianwei has quit IRC | 15:51 | |
| *** mjturek has joined #openstack-ironic | 15:53 | |
| *** r-daneel has quit IRC | 15:54 | |
| * jlvillal assumes lucas-pto is recovering from St Patrick's Day :) | 15:57 | |
| *** gyankum has joined #openstack-ironic | 15:59 | |
| dtantsur | heh | 16:00 |
| *** gyankum has quit IRC | 16:00 | |
| *** namnh has quit IRC | 16:02 | |
| *** pcaruana has quit IRC | 16:07 | |
| *** hjensas has joined #openstack-ironic | 16:09 | |
| *** hjensas has quit IRC | 16:09 | |
| *** hjensas has joined #openstack-ironic | 16:09 | |
| rloo | dtantsur: those inspector-related patches, do they address failures with ironic-tempest-dsvm-ironic-inspector? the etherpad doesn't say much | 16:09 |
| *** mjturek has quit IRC | 16:10 | |
| *** mjturek has joined #openstack-ironic | 16:11 | |
| dtantsur | rloo: they do | 16:13 |
| dtantsur | sorry, detected the issue late on Friday, forgot to update the etherpad | 16:14 |
| rloo | dtantsur: ok, so we need to merge them. otherwise, our gate is broken. | 16:14 |
| rloo | dtantsur: no worries. which patches still need to be reviewed? | 16:14 |
| dtantsur | rloo: we're good as soon as https://review.openstack.org/#/c/553841/ and https://review.openstack.org/#/c/553797/ merge | 16:15 |
| patchbot | patch 553841 - ironic - Gate: stop setting IRONIC_ENABLED_INSPECT_INTEFACE... | 16:15 |
| patchbot | patch 553797 - ironic-inspector - devstack: enable "fake" inspection implementation | 16:15 |
| * TheJulia kind of wonders if slicing/dicing a whiteboard is like cutting onions | 16:15 | |
| TheJulia | https://etherpad.openstack.org/p/IronicWhiteBoard has had lots of updating | 16:15 |
| *** mjturek has quit IRC | 16:15 | |
| jlvillal | I just learned that for Depends-On: we should now use URLs. For example: Depends-On: https://review.openstack.org/536159 | 16:16 |
| patchbot | patch 536159 - openstack-infra/shade - Add devel branches and override-checkout for ansib... (MERGED) | 16:16 |
| rloo | TheJulia: :) Just don't cry. | 16:16 |
| *** fragatina has joined #openstack-ironic | 16:16 | |
| TheJulia | rloo: of course not! sulfuric acid is bad for the eyes! | 16:17 |
| *** skazi has quit IRC | 16:17 | |
| rloo | TheJulia: :D | 16:18 |
| * dtantsur tries to do something refarch-related before the meeting | 16:19 | |
| TheJulia | :) | 16:19 |
| *** crushil has quit IRC | 16:20 | |
| rloo | ok, i updated the whiteboard wrt gate | 16:21 |
| *** mjura has quit IRC | 16:21 | |
| mordred | jlvillal: yah - the new syntax allows folks to do multi-source depends-on - such as an ansible patch depending on a shade-patch - or a patch from one gerrit depending on a patch from a different gerrit | 16:21 |
| dtantsur | what a time to be alive! | 16:21 |
| jlvillal | mordred, cool | 16:22 |
| jlvillal | mordred, Hey I've been having a big discussion about PTI in a patch I added you to :) | 16:23 |
| * TheJulia goes and makes more coffee | 16:23 | |
| mordred | jlvillal: oh fun! looking now | 16:25 |
| jlvillal | mordred, Basically I think that if changing the tox.ini to use "sphinx_build" then should remove the [pbr]/auto_index_modules=True at the same time. | 16:26 |
| jlvillal | mordred, We had a little discussion about it over on #openstack-docs about an hour ago. | 16:27 |
| jlvillal | #openstack-doc not plural | 16:27 |
| *** r-daneel has joined #openstack-ironic | 16:27 | |
| jlvillal | mordred, http://eavesdrop.openstack.org/irclogs/%23openstack-doc/%23openstack-doc.2018-03-19.log.html#t2018-03-19T15:36:59 | 16:28 |
| mordred | jlvillal: yeah ... this one is a fun one - and it occurs to me we might should add a note to the PTI docs about this case | 16:30 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Rework Bare Metal service overview in the install guide https://review.openstack.org/554284 | 16:30 |
| dtantsur | TheJulia, jroll ^^^ | 16:30 |
| jlvillal | mordred, That would be a good idea, IMHO. Thanks! | 16:30 |
| mordred | jlvillal: tl;dr - autodoc_index_modules is a thing we added to pbr ages ago to provide a way to autodoc all of the modules in a project but it's only sporadically used and it adds a pile of complexity that we'd like to see go away | 16:32 |
| mordred | the gate will handle projects that haven't migrated to the new sphinx-build version of the pti because they still need the pbr autodoc | 16:33 |
| mordred | for those projects, updating the tox docs env to run sphinx-build should only be done at the same time as removing the autodoc pbr config (and fixing the doc build to not need it) | 16:34 |
| mordred | otherwise things will be weird | 16:34 |
| *** rpioso|afk is now known as rpioso | 16:35 | |
| jlvillal | mordred, Thanks. Makes sense! | 16:38 |
| *** rwsu has joined #openstack-ironic | 16:38 | |
| sambetts | mordred: is autodoc being replaced by anything? | 16:38 |
| clarkb | sambetts: you can configure autodoc directly in sphinx rather than having pbr do it for you | 16:40 |
| sambetts | clarkb: oh interesting I didn't know that, is that a recent addition to sphinx? we ended up doing this https://github.com/openstack/networking-cisco/blob/master/doc/source/conf.py#L22-L34 in networking cisco to make the pbr autodoc work on read the docs :( | 16:41 |
| *** stendulker has joined #openstack-ironic | 16:42 | |
| clarkb | sambetts: you may have to spell out all the .. automodule:: s? | 16:43 |
| *** trown|ruck is now known as trown|lunch | 16:44 | |
| sambetts | oh :( | 16:46 |
| TheJulia | :( | 16:47 |
| *** hshiina has joined #openstack-ironic | 16:48 | |
| mordred | clarkb: yes. that is the thing that pbr autodoc does - it generates all the automodule statements | 16:48 |
| *** e0ne has quit IRC | 16:48 | |
| rloo | TheJulia: are we ready to update subteam status in whiteboard? /me looks... | 16:49 |
| TheJulia | rloo: update! | 16:49 |
| mordred | sambetts: as of right now nobody is working on a replacement that I know of - although it should be fairly easy to take the pbr logic and extract it to a sphinx plugin - which I'll do at some point but it's currently too far down on my stack | 16:49 |
| rloo | TheJulia: :) doin' it now! | 16:49 |
| TheJulia | update all the things! | 16:49 |
| *** gyee has joined #openstack-ironic | 16:49 | |
| *** hshiina has quit IRC | 16:50 | |
| openstackgerrit | Nguyen Hai proposed openstack/ironic master: Follow the new PTI for document build https://review.openstack.org/553473 | 16:51 |
| johnthetubaguy | rloo: can I sync up about deploy steps and what we do next, I think it was me revising the spec? | 16:51 |
| openstackgerrit | Nguyen Hai proposed openstack/ironic master: Follow the new PTI for document build https://review.openstack.org/553473 | 16:51 |
| *** hshiina has joined #openstack-ironic | 16:51 | |
| *** AlexeyAbashkin has quit IRC | 16:51 | |
| rloo | johnthetubaguy: no. well, i suspect those specs are on hold until after we get the basic framework one done/agreed upon. | 16:52 |
| johnthetubaguy | rloo: I was thinking about the spec for the basic framework really, were you planing on doing that, as I think I said I would make sure it gets done | 16:52 |
| rloo | johnthetubaguy: the only thing we agreed on at the ptg, was to do the deploy steps framework. | 16:52 |
| johnthetubaguy | yeah | 16:52 |
| rloo | johnthetubaguy: oh. i already wrote the spec, but you can work on it too! https://review.openstack.org/#/c/549493/ | 16:53 |
| patchbot | patch 549493 - ironic-specs - Deployment steps framework | 16:53 |
| *** fragatina has quit IRC | 16:53 | |
| rloo | johnthetubaguy: basically a subset of the other spec i had been working on | 16:53 |
| johnthetubaguy | rloo: perfect, I will review that | 16:53 |
| rloo | johnthetubaguy: thx. jroll has some good feedback already. | 16:53 |
| johnthetubaguy | rloo: I was thinking of rebasing the traits thing on top of that | 16:53 |
| rloo | johnthetubaguy: you could but my suggestion is to wait til this is close/approved. | 16:54 |
| rloo | johnthetubaguy: we still have to address/agree on the tougher questions :) | 16:54 |
| johnthetubaguy | rloo: that's fair, its certainly the more important fundamental bit | 16:54 |
| rloo | johnthetubaguy: or maybe that is :-( | 16:54 |
| johnthetubaguy | rloo: I blame the weather | 16:55 |
| *** marios has quit IRC | 16:55 | |
| rloo | johnthetubaguy: the beast from the east. i am sure there will be jokes etc about that beast... | 16:55 |
| johnthetubaguy | rloo: something about it catching a train from denver? | 16:55 |
| rloo | johnthetubaguy: yeah, like that! wondering if there was anything from pre-denver ptg that is memorable/add to the train wreck. | 16:56 |
| * johnthetubaguy scratches head | 16:56 | |
| rloo | for BIOS config -- doesn't anyone know if the code patches are ready for review, or are they still in development? | 16:57 |
| rloo | s/doesn't/does/ | 16:57 |
| *** fragatina has joined #openstack-ironic | 16:58 | |
| TheJulia | rloo: looks like there is a whole list | 16:58 |
| TheJulia | just not an ordered list yet for us to consume | 16:58 |
| TheJulia | zshi: ^^^^ | 16:58 |
| *** jcoufal has joined #openstack-ironic | 16:58 | |
| rloo | TheJulia: yeah, would like someone to comment. until they are sure the patches are good, not sure it is worth starting to review. | 16:59 |
| TheJulia | ditto | 16:59 |
| *** jcoufal_ has quit IRC | 17:00 | |
| openstackgerrit | John Garbutt proposed openstack/ironic-specs master: Deploy Templates https://review.openstack.org/504952 | 17:00 |
| TheJulia | #startmeeting ironic | 17:00 |
| dtantsur | mmmmeeting time? | 17:00 |
| openstack | Meeting started Mon Mar 19 17:00:48 2018 UTC and is due to finish in 60 minutes. The chair is TheJulia. Information about MeetBot at http://wiki.debian.org/MeetBot. | 17:00 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 17:00 |
| TheJulia | o/ | 17:00 |
| *** openstack changes topic to " (Meeting topic: ironic)" | 17:00 | |
| jroll | \o/ | 17:00 |
| dtantsur | o/ | 17:00 |
| openstack | The meeting name has been set to 'ironic' | 17:00 |
| etingof | o/ | 17:00 |
| rpioso | o/ | 17:00 |
| sambetts | o/ | 17:00 |
| rloo | o/ | 17:01 |
| jlvillal | o/ | 17:01 |
| TheJulia | Everyone ready for an ironic meeting?! | 17:01 |
| jlvillal | Uh....yeah...yeah!!! | 17:01 |
| hshiina | o/ | 17:01 |
| TheJulia | Excellent! | 17:01 |
| TheJulia | Our agenda is on the wiki, as always | 17:01 |
| TheJulia | #link https://wiki.openstack.org/wiki/Meetings/Ironic#Agenda_for_next_meeting | 17:01 |
| TheJulia | I made a couple changes this morning. | 17:02 |
| *** hjensas has quit IRC | 17:02 | |
| TheJulia | #topic Announcements / Reminders | 17:02 |
| *** openstack changes topic to "Announcements / Reminders (Meeting topic: ironic)" | 17:02 | |
| *** mjturek has joined #openstack-ironic | 17:02 | |
| mjturek | o/ | 17:02 |
| TheJulia | #info Rocky Cycle Community Priorities have merged https://specs.openstack.org/openstack/ironic-specs/priorities/rocky-priorities.html | 17:02 |
| rloo | yay! | 17:03 |
| TheJulia | #info Regarding the Oneview driver, HPE has indicated they are aware of the requirement to remain in-tree. No other update at this time. | 17:04 |
| *** weshay is now known as weshay_brb | 17:05 | |
| TheJulia | #info XClarity driver CI may be delayed, to be determined. | 17:05 |
| TheJulia | tl;dr Someone new will be picking up the 3rd party CI work, contact unknown at present but I'm in communication with Lenovo. | 17:06 |
| TheJulia | Anyone else have any announcements or reminders to bring up? | 17:06 |
| *** crushil has joined #openstack-ironic | 17:07 | |
| TheJulia | o/ crushil | 17:07 |
| crushil | TheJulia, \o | 17:07 |
| rpioso | Did an ML message go out re: lifecycle management discussions? | 17:07 |
| TheJulia | rpioso: That is on my calendar for thursday | 17:07 |
| rpioso | TheJulia: Cool | 17:08 |
| TheJulia | rpioso: I'll try to get it done sooner :) | 17:08 |
| TheJulia | Well, if there is nothing else announcement wise, we can move on... | 17:08 |
| vdrok | o/ | 17:08 |
| TheJulia | One last thing, the reason for XClarity CI being delayed is that crushil will be moving on. I would like to thank him for his efforts, and hopes that he stays in touch with the community. | 17:10 |
| dtantsur | crushil++ good luck | 17:10 |
| crushil | TheJulia, dtantsur Thanks | 17:10 |
| rloo | thanks crushil, sorry that you're moving on | 17:11 |
| crushil | Thanks rloo. It was pleasure working with the Ironic community | 17:11 |
| TheJulia | Moving on... | 17:11 |
| TheJulia | #topic Reviewing Action Items from the last meeting | 17:11 |
| *** openstack changes topic to "Reviewing Action Items from the last meeting (Meeting topic: ironic)" | 17:11 | |
| TheJulia | We had three action items last week | 17:12 |
| stendulker | o/ | 17:12 |
| TheJulia | (TheJulia) Reorganize the whiteboard | 17:12 |
| TheJulia | (TheJulia) Get the community a taste of storyboard | 17:12 |
| TheJulia | And finally | 17:12 |
| * jlvillal waits... | 17:13 | |
| TheJulia | (mjturek) Leading bug triaging. | 17:13 |
| TheJulia | Sorry, slight interruption | 17:13 |
| jlvillal | heh. I thought it was a dramatic pause :) | 17:13 |
| TheJulia | As everyone has hopefully seen, the whiteboard has been reorganized. | 17:14 |
| TheJulia | #link https://etherpad.openstack.org/p/IronicWhiteBoard | 17:14 |
| TheJulia | I also provided a taste of storyboard with ironic data | 17:14 |
| TheJulia | #link https://nuc.mirroruniverse.org | 17:14 |
| TheJulia | mjturek: how were the bugs? | 17:14 |
| mjturek | did not get a lot done on the triage. Reviewed some of the New bugs but nothing too interesting | 17:14 |
| TheJulia | Sounds good | 17:14 |
| TheJulia | Time to move on to our next topic then! | 17:14 |
| TheJulia | #topic Review subteam status reports | 17:15 |
| *** openstack changes topic to "Review subteam status reports (Meeting topic: ironic)" | 17:15 | |
| TheJulia | #link https://etherpad.openstack.org/p/IronicWhiteBoard | 17:15 |
| TheJulia | Starting at line 148 | 17:16 |
| rloo | sorry this is an aside, but i am looking at the 3rd party CI and it needs to be updated. Maybe an action item for next week TheJulia? L68 | 17:16 |
| TheJulia | Perfectly valid :) | 17:16 |
| rloo | dtantsur: do you know what the critical ironic bug is? | 17:17 |
| TheJulia | #action TheJulia to review/update 3rd party CI section of the whiteboard | 17:17 |
| rloo | dtantsur: forget it, i see it is sushy | 17:17 |
| dtantsur | yep | 17:17 |
| rloo | good work on approving bios config spec. for next week -- would be good to know if the code patches are ready for reviewing. | 17:18 |
| TheJulia | zshi or yolanda, are either of you around? | 17:18 |
| * dtantsur remembers reviewing something | 17:18 | |
| TheJulia | vdrok: Regarding neutron event processing, it looks like the spec likey needs an update. | 17:19 |
| rloo | vdrok: wrt neutron event processing L213. is that the latest status (from today, vs sept)? | 17:19 |
| yolanda | hi TheJulia | 17:19 |
| yolanda | yep, the patches should be ready | 17:19 |
| jroll | awesome | 17:19 |
| *** hjensas has joined #openstack-ironic | 17:19 | |
| *** hjensas has quit IRC | 17:19 | |
| *** hjensas has joined #openstack-ironic | 17:19 | |
| vdrok | rloo: will update the whiteboard, tldr rewriting the code | 17:19 |
| rloo | are we 2018 or 2019? L222... | 17:19 |
| TheJulia | yolanda: we would like someone to create an ordered list for us to begin reviewing from. | 17:19 |
| rloo | thx vdrok! | 17:19 |
| TheJulia | vdrok: thanks! | 17:19 |
| yolanda | ok... let me work on that tomorrow | 17:20 |
| TheJulia | yolanda: thanks, is it okay if I note it as an action item? | 17:20 |
| yolanda | sure | 17:20 |
| openstackgerrit | Michael Turek proposed openstack/ironic master: Add OpenBMC class for accessing nodes using OpenBMC https://review.openstack.org/429836 | 17:20 |
| jroll | TheJulia: the patches are already in order on the branch, right? | 17:20 |
| * jroll sees 5 in one chain | 17:20 | |
| rloo | TheJulia: wrt nova virt/REST API. there were some questions in the priority spec; should we discuss at some point? | 17:20 |
| jroll | rloo: ++ | 17:21 |
| TheJulia | #action yolanda to update list of bios interface patches in the whiteboard to represent review order. | 17:21 |
| *** d0ugal has quit IRC | 17:21 | |
| TheJulia | jroll: rloo: we can discuss it during the discussion or open discussion If that works | 17:22 |
| yolanda | TheJulia, there have been some reviews from Hironori as well. I'll talk with zshi tomorrow and we'll let you know | 17:22 |
| jroll | sure | 17:22 |
| TheJulia | yolanda: thanks | 17:22 |
| yolanda | thx, is great to see progress on that feature! | 17:22 |
| TheJulia | Everyone done reviewing the whiteboard? | 17:22 |
| rloo | TheJulia: wrt storyboard. what is the plan? can we just DO IT? | 17:23 |
| TheJulia | I was going to bring that during discussion :) | 17:23 |
| rloo | TheJulia: okey dokey; i can wait :) | 17:23 |
| TheJulia | short answer yes, long answer, I can have them do it this week and we can start using it say... monday? | 17:23 |
| TheJulia | Anyway, Moving on to priorities for this week | 17:24 |
| TheJulia | #topic Priorities for this next week | 17:24 |
| *** openstack changes topic to "Priorities for this next week (Meeting topic: ironic)" | 17:24 | |
| rloo | fix sushy critical bug | 17:24 |
| dtantsur | yep | 17:25 |
| rloo | dtantsur: you know what needs to be done to ^^ ? | 17:25 |
| dtantsur | rloo: review | 17:25 |
| TheJulia | dtantsur: link? | 17:25 |
| dtantsur | #link https://review.openstack.org/#/c/552817/ | 17:25 |
| patchbot | patch 552817 - sushy - Mark Systems/Managers/SessionService optional | 17:25 |
| rloo | where are we wrt rescue work? | 17:25 |
| TheJulia | Looks like were in disagreement land with regards to interface methods | 17:26 |
| TheJulia | and design... again | 17:26 |
| dtantsur | \o/ | 17:26 |
| rloo | TheJulia: with rescue? | 17:26 |
| TheJulia | rloo: yup | 17:27 |
| TheJulia | and deploy | 17:27 |
| openstackgerrit | Lenny Verkhovsky proposed openstack/ironic master: Infiniband Port Configuration update[1] https://review.openstack.org/553400 | 17:27 |
| jroll | I thought we got rescue done last cycle, except the nova side? | 17:27 |
| rloo | TheJulia: which part/patch of rescue are we talking about, that is in disagreement? | 17:28 |
| jroll | +1, I'm curious | 17:28 |
| dtantsur | adding a new method because for $weird_reason deploy interfaces don't always call boot.prepare_instance.. | 17:29 |
| rloo | TheJulia: is it https://review.openstack.org/#/c/546919/? | 17:29 |
| patchbot | patch 546919 - ironic - Prepare instance to unrescue by deploy interface | 17:29 |
| TheJulia | ^that | 17:29 |
| rloo | ok -- so what needs to be done to move forward. do we need to meet/agree on something? | 17:29 |
| TheJulia | There is some inline discussion, I'd like to keep it as a priority this week to see if we can just get some consensus and move forward. | 17:30 |
| dtantsur | ++ | 17:30 |
| rloo | TheJulia: ok, I'll make a note to look at that. thx. | 17:30 |
| TheJulia | stendulker: In this discussion your patch on fixing for basically fixing the same thing for agent deployments was raised, do you think we can get that revised? | 17:30 |
| TheJulia | It is noted as the "better fix" https://review.openstack.org/#/c/499050/ :) | 17:30 |
| stendulker | TheJulia: I have posted revised patch | 17:30 |
| patchbot | patch 499050 - ironic - Fix ``agent`` deploy interface to call ``boot.prep... | 17:30 |
| TheJulia | Heh, that you have | 17:31 |
| * TheJulia updates priorites list | 17:31 | |
| *** hjensas has quit IRC | 17:31 | |
| stendulker | TheJulia: But one test related to ironic-tempest-dsvm-ipa-wholedisk-agent_ipmitool-tinyipa-multinode fails. Not able to get exact reason for the failure | 17:32 |
| rloo | jroll, others: is that still a critical bug in nova? L112 | 17:32 |
| rloo | jroll: the importance is 'low' in the bug | 17:32 |
| TheJulia | stendulker: Ugh, I can try to take a look this week | 17:32 |
| stendulker | TheJulia: Thanks | 17:33 |
| TheJulia | for nova it's importance may be low, for us it is higher | 17:33 |
| jroll | rloo: I'm not sure I'd call it critical, it's probably high from our perspective (because CI gets into this situation). I updated the patch friday and will poke people today/tomorrow | 17:33 |
| rloo | jroll: ok thx | 17:33 |
| TheJulia | Does anyone have any other items they would like to see reviewed this week? | 17:33 |
| rloo | TheJulia: i've got another question, wrt L117, 'Required Queens Backport'. do we still need that? | 17:33 |
| TheJulia | rloo: I thought I had deleted that... | 17:34 |
| TheJulia | etingof: would you be able to revise https://review.openstack.org/#/c/526773/ in the next day or so? | 17:34 |
| patchbot | patch 526773 - ironic - Add boot mode support to ManagementInterface | 17:34 |
| * etingof looking | 17:35 | |
| rpioso | TheJulia, etingof: I offered to take another look at that patch. | 17:35 |
| etingof | TheJulia, yes, but I'd like to engage rpioso as well | 17:35 |
| TheJulia | Any objection to adding it to the priority list so we can make some progress in that direction? | 17:35 |
| rpioso | +1 | 17:36 |
| etingof | +1! | 17:36 |
| dtantsur | no objections | 17:36 |
| TheJulia | Everyone good with that list of priorites? | 17:36 |
| jroll | I'm happy | 17:37 |
| dtantsur | LGTM | 17:37 |
| rloo | +1 | 17:37 |
| TheJulia | Moving on then! | 17:37 |
| TheJulia | #topic Appointing a bug triaging lead for the coming week | 17:37 |
| *** openstack changes topic to "Appointing a bug triaging lead for the coming week (Meeting topic: ironic)" | 17:37 | |
| TheJulia | Anyone want to be the vict^Wvolunteer this week? | 17:38 |
| *** etingof is now known as etingof|afk | 17:38 | |
| * jroll has to step away for a few, sorry - consider me fine with moving the meeting or keeping it as is when we get to that topic | 17:38 | |
| dtantsur | I can try getting some time for bug triaging | 17:39 |
| *** mjturek has quit IRC | 17:39 | |
| dtantsur | actually, I do some before every meeting anyway | 17:39 |
| rloo | oh, sorry, related to rescue work -- are we planning on doing client release any time soon? | 17:39 |
| TheJulia | If we can get some done before thursday, it will help us out. I'll see about getting them to push the buttons for storyboard migration on Friday | 17:39 |
| rloo | thx dtantsur! | 17:39 |
| TheJulia | rloo: already done | 17:39 |
| rloo | TheJulia: sweet. thx! | 17:39 |
| TheJulia | no problem | 17:40 |
| TheJulia | Seems like we should all try to spend time on bugs this week... | 17:40 |
| *** mjturek has joined #openstack-ironic | 17:40 | |
| TheJulia | Anyway, I'm going to move on to discussion since we really all should try and spend some time looking at launchpad bugs this week | 17:41 |
| TheJulia | #topic Discussion | 17:41 |
| *** openstack changes topic to "Discussion (Meeting topic: ironic)" | 17:41 | |
| TheJulia | We have two items... well three, maybe. :) | 17:41 |
| TheJulia | The first is do we move the meeting to 2PM or 3PM UTC? | 17:42 |
| TheJulia | The results from the time poll can be found at https://doodle.com/poll/6kuwixpkkhbwsibk | 17:42 |
| *** etingof|afk has quit IRC | 17:42 | |
| dtantsur | this is great for Europe, but I guess we exclude west coast? | 17:43 |
| TheJulia | And those times are friendly-er to Japan in that it may be 11PM or 12PM. We don't have any New Zealand contributors, but it would also mean that the west coast of the united states would have to get up early :) | 17:43 |
| TheJulia | I am willing to get up for a 7AM meeting, because I <3 you all | 17:43 |
| dtantsur | like, really early :) | 17:43 |
| TheJulia | Eh, not that early | 17:44 |
| jlvillal | https://www.timeanddate.com/worldclock/fixedtime.html?msg=Ironic+meeting&iso=20180327T02&p1=1440&ah=1 | 17:44 |
| jlvillal | Oh wait wrong | 17:44 |
| jlvillal | https://www.timeanddate.com/worldclock/fixedtime.html?msg=Ironic+meeting&iso=20180327T14&p1=1440&ah=1 | 17:45 |
| *** hjensas has joined #openstack-ironic | 17:45 | |
| *** hjensas has quit IRC | 17:45 | |
| *** hjensas has joined #openstack-ironic | 17:45 | |
| TheJulia | I think the time should also be a little more friendly to stendulker | 17:45 |
| jlvillal | So 7am here on the west coast of the USA | 17:45 |
| TheJulia | Thoughts/objections? Shall we just vote on it? | 17:45 |
| stendulker | TheJulia: yes | 17:45 |
| * dtantsur cannot seriously ask people to wake up at 7am, which is deep night for him | 17:45 | |
| rpioso | jlvillal: That's during daylight savings time. It'll change in the fall. | 17:46 |
| jlvillal | So 2PM UTC is 7AM pacific now. When Daylight Savings Times end it will be 6AM | 17:46 |
| TheJulia | It is worth noting that I don't think we have any active contributors in .au or .nz who the time would be the most unfriendly to | 17:46 |
| TheJulia | I like 6AM less than 7 AM | 17:46 |
| TheJulia | Perhaps we vote on 3PM UTC or not changing then? | 17:46 |
| jlvillal | 3PM times: https://www.timeanddate.com/worldclock/fixedtime.html?msg=Ironic+meeting&iso=20180327T15&p1=1440&ah=1 | 17:47 |
| jlvillal | Find your location :) | 17:47 |
| * TheJulia finds it amusing that the website knows I'm in Palm Springs, CA area | 17:47 | |
| rloo | so based on the poll, we are talking about 2pm vs 3pm, right? the only diff is jlvillal or hjensas if I read it correctly. | 17:48 |
| rloo | and any future west coast US folks. | 17:48 |
| rloo | so between the two times, much as I'd like hjensas, it seems like 3pm is the saner one? | 17:48 |
| dtantsur | it feels so... | 17:48 |
| dtantsur | so, essentially, move 1 hour earlier? | 17:49 |
| jlvillal | Two hours | 17:49 |
| rloo | is hjensas here? | 17:49 |
| TheJulia | Yes, two hours, we are presently 5PM UTC | 17:50 |
| rloo | (I guess i am wondering if hjensas would like to attend the meeting if he could attend the meeting) | 17:50 |
| *** weshay_brb is now known as weshay | 17:50 | |
| dtantsur | rly? | 17:50 |
| dtantsur | hmm, maybe you're right /me hates timezones | 17:50 |
| jlvillal | dtantsur, https://www.timeanddate.com/worldclock/fixedtime.html?msg=Ironic+meeting&iso=20180327T15&p1=1440&ah=1 Find Berlin :) | 17:50 |
| dtantsur | jlvillal: this link is wrong, that's the problem.. Berlin is not UTC+2 | 17:51 |
| dtantsur | that's what confused me | 17:51 |
| jlvillal | dtantsur, Ah, so not 7PM-ish right now where you are? | 17:51 |
| TheJulia | I know this meeting is presently ?2 AM? for hshiina, so if we would like to vote we can now, or we can defer until next week. | 17:51 |
| dtantsur | ah, 27 Mar will be after daylight switch | 17:51 |
| TheJulia | dtantsur: I think we all dislike time zones | 17:51 |
| jlvillal | dtantsur, Sorry I made it for next week :( | 17:52 |
| dtantsur | I think we can vote, his vote is on the poll | 17:52 |
| * TheJulia tries to remember voting syntax | 17:52 | |
| jlvillal | #vote | 17:52 |
| TheJulia | #vote 3pm nochange | 17:53 |
| * jlvillal is ignored as not meeting operator | 17:53 | |
| dtantsur | TheJulia: #startvote Description Options | 17:53 |
| dtantsur | or maybe even #startvote Description? Options | 17:53 |
| jlvillal | yes no. I assume for options | 17:53 |
| jlvillal | #help startvote | 17:53 |
| rloo | So the choices are 2pm or 3pm, right? | 17:53 |
| TheJulia | #startvote Move_The_Meeting 3PM KeepTheSame | 17:53 |
| dtantsur | #startvote Should we move this meeting to 3pm UTC? Yes, No | 17:53 |
| openstack | Unable to parse vote topic and options. | 17:53 |
| openstack | Only the meeting chair may start a vote. | 17:53 |
| jlvillal | meh. Maybe I should submit a patch... | 17:53 |
| dtantsur | TheJulia: copy my message | 17:53 |
| TheJulia | #startvote Should we move this meeting to 3pm UTC? Yes, No | 17:54 |
| openstack | Begin voting on: Should we move this meeting to 3pm UTC? Valid vote options are Yes, No. | 17:54 |
| openstack | Vote using '#vote OPTION'. Only your last vote counts. | 17:54 |
| TheJulia | \o/ | 17:54 |
| dtantsur | :) | 17:54 |
| TheJulia | #vote Yes | 17:54 |
| dtantsur | #vote Yes | 17:54 |
| rloo | if we say no, then 2pm or we keep it at 5pm? | 17:54 |
| sambetts | #vote Yes | 17:54 |
| TheJulia | no would keep it at this point | 17:54 |
| rpioso | #vote Yes | 17:54 |
| rloo | #vote Yes | 17:54 |
| rloo | i think jroll would also vote yes | 17:54 |
| hshiina | #vote Yes | 17:54 |
| *** jcoufal has quit IRC | 17:55 | |
| stendulker | #vote Yes | 17:55 |
| TheJulia | Say I'll close voting in about 1 minute | 17:55 |
| *** jcoufal has joined #openstack-ironic | 17:56 | |
| jlvillal | #vote yes | 17:56 |
| TheJulia | Our next discussion topic is storyboard. If there are no objections, I'll tell infra to push the buttons later this week and we can start using storyboard next monday. There will naturally be an adjustment period, and we may want to have a call to discuss use/patterns next week. | 17:56 |
| TheJulia | #endvote | 17:56 |
| openstack | Voted on "Should we move this meeting to 3pm UTC?" Results are | 17:56 |
| openstack | Yes (8): TheJulia, rloo, jlvillal, stendulker, rpioso, dtantsur, sambetts, hshiina | 17:56 |
| TheJulia | So, I guess we move the meeting | 17:56 |
| dtantsur | yep | 17:56 |
| hjensas | Sorry, I just came in. But 3pm UTC is good for me. :) | 17:56 |
| TheJulia | #action TheJulia to reschedule and send out email. | 17:56 |
| TheJulia | Awesome, thanks hjensas | 17:57 |
| TheJulia | Any objections/issues/questions regarding storyboard? | 17:57 |
| jlvillal | TheJulia, So that means you will submit patch? | 17:57 |
| rloo | hjensas: great | 17:57 |
| TheJulia | jlvillal: Yes | 17:57 |
| jlvillal | Thanks! | 17:57 |
| TheJulia | rloo: jroll: I guess we can discuss nova virt driver stuff in channel when jroll returns. | 17:57 |
| rloo | TheJulia: yup | 17:57 |
| TheJulia | 3 Minute Warning | 17:57 |
| dtantsur | anyway 3 minutes is not enough for a nova discussion :) | 17:58 |
| rloo | who else is interested in the nova virt discussion besides dtantsur, TheJulia, jroll, me? | 17:58 |
| dtantsur | I will have to leave soon | 17:58 |
| rloo | adn wrt storyboard, action item TheJulia? | 17:58 |
| dtantsur | yeah, and #agreed on the meeting :) | 17:58 |
| rloo | dtantsur: so maybe tomorrow instead, to talk about nova virt | 17:58 |
| TheJulia | #agreed Moving meeting to 3pm UTC | 17:59 |
| dtantsur | don't wait for me, I guess | 17:59 |
| TheJulia | #action TheJulia to start storyboard migration | 17:59 |
| jroll | tomorrow wfm, sorry about leaving abruptly | 17:59 |
| dtantsur | I'll read the scrollback | 17:59 |
| rloo | TheJulia: do we need to hold off opening/updating bugs during xsition to storyboard? | 17:59 |
| TheJulia | rloo: Only on friday I guess | 17:59 |
| TheJulia | I'll follow-up | 18:00 |
| rloo | TheJulia: ok, just give us some notice :) | 18:00 |
| TheJulia | Thanks everyone! | 18:00 |
| dtantsur | thanks TheJulia | 18:00 |
| TheJulia | rloo: will do! | 18:00 |
| TheJulia | #endmeeting | 18:00 |
| *** openstack changes topic to "Bare Metal Provisioning | Status: http://bit.ly/ironic-whiteboard | Docs: http://docs.openstack.org/developer/ironic/ | Bugs: https://bugs.launchpad.net/ironic" | 18:00 | |
| openstack | Meeting ended Mon Mar 19 18:00:17 2018 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 18:00 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/ironic/2018/ironic.2018-03-19-17.00.html | 18:00 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/ironic/2018/ironic.2018-03-19-17.00.txt | 18:00 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/ironic/2018/ironic.2018-03-19-17.00.log.html | 18:00 |
| TheJulia | dtantsur: jroll: rloo: okay, tomorrow morning US timeframe once we're all awake then? I don't want to move forward on this sooner rather than later this cycle because we had easy consensus from nova and don't want the opinions to shift again | 18:01 |
| dtantsur | yep | 18:02 |
| jroll | TheJulia: I'm free til 1300 EST | 18:02 |
| jroll | and agree | 18:02 |
| rloo | TheJulia: agree on both counts. tomorrow. everyone is invited! :) | 18:02 |
| TheJulia | So was the concern why were not just hacking support to do the explicit microversion calls into python-ironicclient? | 18:02 |
| dtantsur | I guess we'll have to do it eventually | 18:03 |
| openstackgerrit | Merged openstack/ironic master: Gate: stop setting IRONIC_ENABLED_INSPECT_INTEFACES=inspector https://review.openstack.org/553841 | 18:03 |
| openstackgerrit | Merged openstack/ironic-inspector master: devstack: enable "fake" inspection implementation https://review.openstack.org/553797 | 18:03 |
| TheJulia | I kind of don't think we'll have to | 18:03 |
| TheJulia | the object based mocking just makes it easy to break interactions and it hides a lot.... and we have a lot of code that is not needed that does special things | 18:04 |
| TheJulia | Also there is the whole growing consensus to stop using the client libraries for service to service communication because it turns out that it actually makes it harder | 18:04 |
| dtantsur | the gate should be back to norm now | 18:04 |
| TheJulia | \o/ | 18:04 |
| pas-ha | heya, new setuptools! | 18:06 |
| pas-ha | and it broke nova and neutron on Pike (at least) | 18:06 |
| pas-ha | we should check our tests as well | 18:06 |
| dtantsur | ugh | 18:06 |
| dtantsur | folks, two simple patches for your review stats: https://review.openstack.org/#/c/543989/ and https://review.openstack.org/#/c/548971/ | 18:06 |
| patchbot | patch 543989 - ironic-tempest-plugin - Add a full set of tests for the IPMI hardware type... | 18:06 |
| jroll | TheJulia: so I agree with you, my thoughts are: 1) I think application developers using the ironic sdk will eventually want per-call microversion support, and 2) the less code we churn, the quicker we can get this done and keep moving. I'm hoping we can get the per-call microversion things done, and then do the rest as a refactor | 18:06 |
| patchbot | patch 548971 - ironic-inspector - Correct Queens release notes versions | 18:07 |
| pas-ha | the broken part is common though - version comparison in oslo.versionedobjects | 18:07 |
| *** mjturek has quit IRC | 18:07 | |
| jroll | TheJulia: I'm curious if you have some links or something that show this growing consensus about this, I personally haven't seen much of that | 18:07 |
| *** tesseract has quit IRC | 18:07 | |
| TheJulia | jroll: Largely hallway track discussion combined with denver api-sig discussion. | 18:08 |
| pas-ha | fwiw setuptools 39.0.0 started process of moving pkg_resources to a separate package... and it seems this does not work as expected anymore | 18:08 |
| TheJulia | dtantsur: Do you remember where those discussions ended up at? | 18:08 |
| dtantsur | I don't, unfortunately | 18:08 |
| jroll | gotcha. | 18:09 |
| dtantsur | see you tomorrow | 18:09 |
| TheJulia | I do very distinctly remember that it was brought up, and that there was consensus from the api-sig. Also there was a call to NOT use openstacksdk for service to service communication in dublin | 18:09 |
| *** dtantsur is now known as dtantsur|afk | 18:09 | |
| dtantsur|afk | o/ | 18:09 |
| TheJulia | goodnight dtantsur|afk | 18:09 |
| jroll | I do recall the latter | 18:09 |
| TheJulia | jroll: I guess I'm not afraid of just doing rest calls with the session client directly at this point because we did it with shade... Rewriting a ton of unit tests and finding broken tests built upon not great assumptions about behavior is going to be a bit more of a headache. | 18:11 |
| rloo | pas-ha: :-(. We fix one thing, then something else crops up... | 18:11 |
| * sambetts --> home | 18:12 | |
| sambetts | night all o/ | 18:12 |
| TheJulia | goodnight sambetts | 18:12 |
| *** sambetts is now known as sambetts|afk | 18:12 | |
| rloo | bye dtantsur|afk, sambetts|afk | 18:12 |
| jroll | TheJulia: yeah, my main thing is that the less code we have to change in nova, the quicker we can get it done. we can get the primary goal done much quicker and then refactor the rest as we go | 18:13 |
| TheJulia | I think the clean refactor has higher value, even if it is more code. Doing it in two steps should also make it an easier review even if it is a lot of code for the initial testing changes. :\ | 18:15 |
| TheJulia | melwitt: thoughts^ | 18:15 |
| jroll | ok | 18:15 |
| jroll | we're low on people-time so I have to ask :) | 18:15 |
| TheJulia | regardless of what we do, with the object based mocking, the testa re going to have to get blown up and rewritten | 18:16 |
| rloo | pas-ha: you think this is broken cuz of what you mentioned? https://review.openstack.org/#/c/545297/ | 18:16 |
| patchbot | patch 545297 - ironic-tempest-plugin - Validate scheduling fields in basic ops scenario | 18:16 |
| jroll | I don't think the tests would change much if we just add the microversion to the call, honestly | 18:16 |
| rloo | pas-ha: not the inspector one, the pike ones | 18:16 |
| TheJulia | s/testa\ /tests a/ | 18:16 |
| TheJulia | jroll: that is correct, but to actually do the right thing and an eventual refactor is still a complete partial detonation of the tests | 18:17 |
| TheJulia | on a plus side, we'll find all the testing assumptions that are not great if we do it | 18:17 |
| jroll | yes, but it's easier to say "we'll definitely ship this in rocky" | 18:17 |
| jroll | yep | 18:17 |
| TheJulia | I'll try and discuss with melwitt and we can kind of go from there. | 18:18 |
| * melwitt is reading | 18:18 | |
| jroll | nod | 18:18 |
| TheJulia | because I do agree with you, the whole idea is to stop bandaiding and shooting ourselves in the foot | 18:18 |
| melwitt | are you discussing whether or not to refactor the tests in a first step change then put the REST call stuff on top? | 18:18 |
| jroll | melwitt: the question is should we do that, or, should we continue using the object-based stuff, add per-call microversions, and then later refactor to straight REST | 18:19 |
| TheJulia | melwitt: discussing if it is valuable enough to do. We have no real choice with how the tests are written but to do that, so it would be a bulk test rewrite in a single patch or two, then we could piecemeal routes of the code to be rest based | 18:19 |
| TheJulia | melwitt: I have a wip you can look at if you want to cry | 18:20 |
| TheJulia | :) | 18:20 |
| melwitt | heh | 18:20 |
| * jroll thinks we're on two different topics, maybe :/ | 18:20 | |
| *** AlexeyAbashkin has joined #openstack-ironic | 18:20 | |
| *** stendulker has quit IRC | 18:20 | |
| TheJulia | jroll: same topic, two different angles | 18:20 |
| melwitt | jroll: I thought we couldn't add per-call microversions to the object-based stuff bc ironicclient doesn't let us? | 18:20 |
| jroll | melwitt: well yes, we'd fix ironicclient first :) | 18:20 |
| TheJulia | we would have to wire it completely through, we would keep the dependency | 18:20 |
| TheJulia | the assumption during the meeting is that we had already gutted our special logic and were just a keystoneauth1 session wrapper | 18:21 |
| TheJulia | which is not the case | 18:21 |
| melwitt | yeah, okay. fixing ironicclient would be the ideal state, right? and the REST call stuff was just to bridge the gap until that happens (or just leave it as converting back to ironicclient would be a pain) | 18:21 |
| TheJulia | melwitt: reverse that | 18:21 |
| openstackgerrit | Nguyen Hai proposed openstack/python-ironicclient master: Follow the new PTI for document build https://review.openstack.org/554157 | 18:22 |
| TheJulia | There is the whole desire to do away with client libraries so restification becoes ideal | 18:22 |
| melwitt | I see | 18:22 |
| TheJulia | wiring in microversion specific call logic and try/expect catching woudl be the workaround for now I guess | 18:23 |
| TheJulia | since there was huge resistance to if version x do y logic.... | 18:23 |
| * TheJulia doesn't understand why there was that resistance when discussing in dublin | 18:23 | |
| *** AlexeyAbashkin has quit IRC | 18:24 | |
| * TheJulia feels like we needed more than an hour... and Guinness to be delivered while discussing. | 18:25 | |
| *** trown|lunch is now known as trown | 18:25 | |
| melwitt | I don't think I caught the details of that. you're suggesting querying ironic version at startup and then "if version X, do Y, else do Z?" | 18:25 |
| *** trown is now known as trown|compassing | 18:25 | |
| TheJulia | melwitt: we built in revised negotiation logic last cycle to do the if version x, do y else z logic based upon a list of "supported versions" | 18:26 |
| TheJulia | although api-sig guidance is now going towards per call microversioning | 18:26 |
| *** hshiina has quit IRC | 18:27 | |
| *** anupn has left #openstack-ironic | 18:27 | |
| jroll | I'm apathetic to both what order we do the code in, and how the logic works, fwiw | 18:28 |
| melwitt | TheJulia: sorry, I'm afraid I'm not following. you're saying last cycle in the ironic driver logic was added to do the "if version x" thing already? then what are we trying to solve this cycle? | 18:28 |
| jroll | I just want to make sure we get it done, and wanted to ask the question if it's worth doing the rest thing now or wait for later | 18:29 |
| TheJulia | melwitt: sorry, in python-ironicclient we added logic to help expose what the remote server was speaking version wise | 18:29 |
| TheJulia | what we're trying to solve this cycle is the horrible habit of a singular microversion pin that breaks upgrade testing... ever... single... cycle | 18:29 |
| TheJulia | well, not direclty, but it prevents us from being able to route aroudn it so we're in a situation were we have to make a job non-voting just to make progress on completely unrelated areas until we get our grenade job working again | 18:30 |
| *** rama_y has joined #openstack-ironic | 18:31 | |
| melwitt | okay, so the choices are, 1) leverage the version exposure from ironicclient in the ironic driver to behave differently based on version, 2) "fix" ironicclient to do per-call microversions, 3) raw REST calls | 18:31 |
| TheJulia | So going back to discussion around Denver timeframe was that we would create a list of known supported microversions, update the ironic virt driver to know about that list. That list would be sent to python-ironicclient which would then negotiate the higest available version, and expose a property (which is done) | 18:31 |
| TheJulia | that property could have then been used for if x version do y, else z | 18:31 |
| TheJulia | melwitt: basically, yes and that is also in order of easy to difficult | 18:32 |
| pas-ha | rloo: does not seems so, until now we've seen only broken unit tests | 18:33 |
| pas-ha | (all related to versioned objects) | 18:34 |
| *** serlex has joined #openstack-ironic | 18:34 | |
| rloo | pas-ha: ok. then something else is broke.n. maybe. | 18:34 |
| *** serlex has quit IRC | 18:34 | |
| pas-ha | might be :-( | 18:34 |
| rloo | pas-ha: i'll wait a bit to see if things 'settle down'... | 18:34 |
| melwitt | TheJulia: yeah. it seems like 2) would be the most practical option, IMHO. and as jroll mentioned earlier would have utility beyond this particular issue (allowing client users to request specific microversions) | 18:35 |
| pas-ha | rloo: on the other hand there is this http://logs.openstack.org/97/545297/10/check/ironic-dsvm-standalone-pike/33c79bb/logs/screen-q-agt.txt.gz | 18:35 |
| melwitt | *per-call | 18:36 |
| *** e0ne has joined #openstack-ironic | 18:36 | |
| rloo | pas-ha: ouch. | 18:36 |
| TheJulia | there is another issue with per-call being problematic | 18:37 |
| * TheJulia goes and finds lunch to try and remember | 18:38 | |
| *** milan_ has quit IRC | 18:38 | |
| *** Goneri has quit IRC | 18:41 | |
| *** openstackgerrit has quit IRC | 18:48 | |
| *** openstackgerrit has joined #openstack-ironic | 18:52 | |
| openstackgerrit | Nguyen Hai proposed openstack/ironic master: Follow the new PTI for document build https://review.openstack.org/553473 | 18:52 |
| openstackgerrit | Merged openstack/sushy-tools master: Abstract away the virtualization management interface https://review.openstack.org/523321 | 18:57 |
| *** krtaylor has joined #openstack-ironic | 18:59 | |
| TheJulia | melwitt: So I think that belief that per-call couldn't be done came out of a belief that sessionclient just gave it to us for free over this past cycle, and that in reality ironic had a lot of code between the the entry into the client and the http request going out the back-end of the client. It looks like it could be done with work on python-ironicclient where we could wire it all in to all the individual method | 19:04 |
| TheJulia | calls. The downside is we would need to either abandon the default microversion behavior upon connection/initial negotiation, or treat it as an override, or keep the "pin" but use latest" which allows the client to try to speak the most highest known version. I'm not sure how well IRONIC_API_VERSION="latest" and versions on each call would be received by nova reviewers. | 19:04 |
| * TheJulia wonders if we just bump the minor and remove the hard pin internally | 19:06 | |
| jroll | has anyone noticed that our specs pages look (IMHO) terrible with the openstackdocstheme that we switched to last week? e.g. http://specs.openstack.org/openstack/ironic-specs/specs/9.0/boot-from-volume-reference-drivers.html | 19:09 |
| * TheJulia blinks | 19:10 | |
| TheJulia | Not pretty on a mobile device :( | 19:10 |
| jroll | even on a computer :( | 19:10 |
| jroll | the commit message and bug don't say why the switch, though oslosphinx's README implies it's dead | 19:11 |
| melwitt | TheJulia: yeah, not sure what the best practice is there. dansmith, do you happen to know what would be recommended for ironicclient version (if per-call microversion support is added) when initializing the client object upon startup? | 19:13 |
| *** zshi has quit IRC | 19:13 | |
| dansmith | melwitt: I'm not sure what you're asking.. if it's per-call, then you don't need to initialize it at startup right? | 19:14 |
| dansmith | I mean, you could look for the max of all the minimums across all the calls and make sure the server has at least that value I guess | 19:14 |
| TheJulia | thats a good point, resource tracking will still cause the client to connect | 19:15 |
| dansmith | but, we don't do that in, for example, novaclient AFAIK | 19:15 |
| melwitt | that's what I wasn't sure about. I think when you initialize the client object, you have to pass a version also, maybe to act as the default. at least it looks like novaclient python API works that way | 19:15 |
| TheJulia | we would still need to gut our default out | 19:15 |
| dansmith | I would just check for some sane minimum if you want, and then fail per-call if you can't do something because of a versions | 19:15 |
| dansmith | *version | 19:15 |
| TheJulia | or disregard all of that logic | 19:15 |
| dansmith | if you version per call, there's really no default | 19:15 |
| TheJulia | well, that would require all calls to be called with their specific default | 19:16 |
| TheJulia | a global overarching default though would still exist | 19:16 |
| dansmith | it just depends on how much you hide in the client | 19:17 |
| melwitt | underneath yeah, there will be a default because AFAIK you can't not have one. just if you specify on every call, it'll never fall back to the default | 19:18 |
| TheJulia | That is what I'm thinking | 19:18 |
| dansmith | didn't you say you were going to convert to raw requests for this stuff? | 19:18 |
| jroll | oh my, what did I start | 19:19 |
| dansmith | or is this that the keystone session has to have something for the default? | 19:19 |
| melwitt | we were talking through the options because the raw request stuff will be a very large change | 19:19 |
| TheJulia | basically the unit tests would be a complete initial rewrite | 19:20 |
| openstackgerrit | Aleksandra Bezborodova proposed openstack/sushy-tools master: Add unit tests for emulator https://review.openstack.org/554326 | 19:20 |
| *** AlexeyAbashkin has joined #openstack-ironic | 19:20 | |
| TheJulia | I say that because we would want unit tests that could tell us if the new rest statements are working as expected and presently we mock ironicclient completely so all behavior is masked | 19:22 |
| melwitt | yeah. so we were thinking maybe ironicclient per-call microversion support might be a good stepping stone and the raw request refactor could be a later thing to take more time on | 19:23 |
| dansmith | you don't really need to pass a microversion to ironicclient then do you? | 19:24 |
| *** fragatina has quit IRC | 19:24 | |
| TheJulia | dansmith: what do you mean? | 19:24 |
| dansmith | if you have abstract methods you just call with arguments, it knows what version(s) it supports for that call, and tries to do them against the server, or fails if you ask for something it can't do? | 19:24 |
| *** AlexeyAbashkin has quit IRC | 19:24 | |
| TheJulia | That would be a maintenance nightmare for something intended to be lightweight | 19:25 |
| *** zshi has joined #openstack-ironic | 19:26 | |
| dansmith | um, okay | 19:26 |
| TheJulia | I thought, in the discussion at least my perception of the discusison in dublin was that visibility was desired by nova | 19:26 |
| dansmith | sounds like ya'll don't need my help | 19:26 |
| jroll | TheJulia: I think what dansmith is saying is put the version handling logic in the thing doing the REST call (whether that's in nova or ironicclient), assuming that we do the try/except catching logic and not the if/else logic | 19:30 |
| jroll | either way you have to know the version and fail if you can't, just a matter of where that code lives | 19:30 |
| * jroll hopes that makes sense, words are hard | 19:30 | |
| melwitt | isn't TheJulia saying that ironicclient already does that? and the decision point is whether or not it's "okay" for ironic driver to react differently based on the return/error from ironicclient? | 19:31 |
| TheJulia | jroll: I guess my frustration there is we've gone completely full circle with the slight change now, we declare it in the caller and hope the remote side has a sufficent version | 19:31 |
| jroll | gotcha. | 19:32 |
| TheJulia | melwitt: Essentially yes, our support policy is 1.1->whatever our latest version is at present, if we start ripping out stuff... well.. then it gets complicated | 19:32 |
| *** crushil has quit IRC | 19:32 | |
| TheJulia | and that actually still wouldn't solve the problem as to why we ended up being pinned on version 1.9 | 19:33 |
| TheJulia | The method did not change, but the resulting behavior changed and the community decision was to attempt to mask the breaking change as much as possible | 19:33 |
| *** dprince has quit IRC | 19:33 | |
| * TheJulia feels like beer is required | 19:33 | |
| jroll | TheJulia: like I said, I'm fine however we get it done, I'm just trying to understand if the options are equally good, and if so, which is the path of least resistance. AIUI at the moment, pure rest is the better solution, and at least one person (you) thinks that we can still accomplish that this cycle. so unless we have a reason not to do that, let's go that route | 19:34 |
| *** e0ne has quit IRC | 19:34 | |
| TheJulia | jroll: you've not realized I'm super optimistic.... right? | 19:35 |
| melwitt | I'm concerned if there's a misunderstanding blocking the easiest solution 1) we mentioned earlier | 19:35 |
| melwitt | tbh, it's not clear to me if/why anyone opposes it | 19:35 |
| TheJulia | I feel like the opposition to easy is the desire for perfection | 19:36 |
| TheJulia | I'm fine with easy, in fact I prefer it | 19:36 |
| TheJulia | as long as it makes sense and doesn't shoot us in the foot | 19:36 |
| jroll | TheJulia: about getting it done? I didn't get that from this conversation. if we don't think we can do (3) this cycle, then I think we should aim for (2) | 19:36 |
| TheJulia | jroll: I was super confident it could be done in like a week until I started digging in the virt driver unit tests last week and lost a day to what was actually bad test assumptions | 19:37 |
| jroll | my problem with (1) is it requires instantiating a new ironicclient instance to go try a different microversion, which is fine, just feels hacky (and makes calls to keystone and such) | 19:37 |
| TheJulia | jroll: we now support downgrading the request without a new client... | 19:37 |
| jroll | orly | 19:38 |
| TheJulia | but only for a list of versions | 19:38 |
| TheJulia | yarly | 19:38 |
| jroll | ah, we pass versions to try? | 19:38 |
| TheJulia | we can pass a list to the os_ironic_api_version now | 19:38 |
| jroll | gotcha | 19:38 |
| jroll | that's... interesting, but if you need to change the arguments (e.g. decide whether to pass configdrive to rebuild), it doesn't help, right | 19:38 |
| TheJulia | and should we somehow hit an even older version, it should then downgrade further | 19:39 |
| TheJulia | jroll: that is why we added the ability to get the version out of the existing client | 19:39 |
| jroll | nod | 19:39 |
| jroll | so it does seem we have a really easy path, then, we just want 'better' | 19:40 |
| TheJulia | It wouldn't be hard to provide a reset mechanism to that, but the cleaner thing would be a kwarg of microversion that we pickup on and act upon | 19:40 |
| jroll | agree | 19:40 |
| jroll | so we have done, better, and best options, in order of amount of work | 19:41 |
| TheJulia | that is a good way of putting it, done seemed to be not liked, so better seems logical. I guess I'll start hacking on that in the morning | 19:42 |
| *** deadnull has joined #openstack-ironic | 19:42 | |
| jroll | if you think best is right, don't let me stop you, I'm just waving my hands around and trying to be pragmatic | 19:43 |
| jroll | :) | 19:43 |
| jroll | at any rate, thank you for taking this on | 19:43 |
| melwitt | TheJulia, jroll: fwiw, I think it would be fine to do option 1) for the version stuff we discussed earlier. if ironicclient has version negotiation built in and will return/error based on what it tried to do, ironic driver can react to that and do the right thing | 19:44 |
| jroll | right on, thanks melwitt. I'm also cool with that | 19:45 |
| melwitt | the thing people were opposed to was, the suggestion to change the upgrade order to "solve the problem" | 19:45 |
| melwitt | which is totally not related to option 1) | 19:45 |
| jroll | ++ | 19:45 |
| *** openstackgerrit has quit IRC | 19:48 | |
| TheJulia | I would absolutely love to be able to change the order as the underlying openstack tooling for upgrade testing is no longer cutting it and anything to make the pain less would be appreciated. But at the same time, at least offering the ability to be a little more tolerant seems reasonable (which amazingly enough, there was agreement on), as long as we don't change the default testing order. | 19:50 |
| TheJulia | The reality, as I understand it, of changing that upgrade order, is it is never going to happen short of replacing grenade. | 19:51 |
| melwitt | yeah, to be clear, we don't want to change the upgrade order. but option 1) for updating ironic driver to handle the already supported ironicclient version negotiation is cool | 19:51 |
| jroll | I don't think everyone is opposed to changing the upgrade order, I just think they want ops to have lots of notice and noise before it happens. or at least that's the camp I'm in. | 19:52 |
| melwitt | ack | 19:52 |
| TheJulia | jroll: agreed, and really it should be moving forward as much about upgrade order, but declared interoperability between component versions. At least in a perfect universe... which might exist in the theoretical multiverse. | 19:53 |
| jroll | :) | 19:54 |
| *** slagle has quit IRC | 19:56 | |
| *** milan has joined #openstack-ironic | 19:56 | |
| TheJulia | Someone posted a really funny clip of what each MCU/Xmen character would say upon meeting and Beast says he created a device to cross the multiverse... and the very last line is deadpool saying "Fox and Disney made a deal" | 19:57 |
| TheJulia | s/clip/picture/ | 19:57 |
| *** slagle has joined #openstack-ironic | 19:58 | |
| *** deadnull has quit IRC | 20:03 | |
| jroll | heh | 20:05 |
| *** jcoufal has quit IRC | 20:18 | |
| *** AlexeyAbashkin has joined #openstack-ironic | 20:21 | |
| *** openstackgerrit has joined #openstack-ironic | 20:22 | |
| openstackgerrit | Nguyen Hai proposed openstack/python-ironicclient master: Follow the new PTI for document build https://review.openstack.org/554157 | 20:22 |
| *** jcoufal has joined #openstack-ironic | 20:23 | |
| *** Sukhdev_ has joined #openstack-ironic | 20:25 | |
| *** livelace has joined #openstack-ironic | 20:25 | |
| *** AlexeyAbashkin has quit IRC | 20:26 | |
| *** jcoufal_ has joined #openstack-ironic | 20:33 | |
| *** jcoufal has quit IRC | 20:35 | |
| *** jcoufal_ has quit IRC | 20:38 | |
| *** trown|compassing is now known as trown|ruck | 20:43 | |
| *** Goneri has joined #openstack-ironic | 20:43 | |
| *** fragatina has joined #openstack-ironic | 20:52 | |
| *** jcoufal has joined #openstack-ironic | 20:54 | |
| *** anupn has joined #openstack-ironic | 20:59 | |
| *** jtomasek has quit IRC | 21:00 | |
| *** trown|ruck is now known as trown|outtypewww | 21:01 | |
| *** itlinux has quit IRC | 21:06 | |
| *** livelace has quit IRC | 21:08 | |
| *** e0ne has joined #openstack-ironic | 21:09 | |
| *** mgoddard has quit IRC | 21:14 | |
| *** jcoufal_ has joined #openstack-ironic | 21:18 | |
| *** jcoufal has quit IRC | 21:18 | |
| *** jcoufal_ has quit IRC | 21:28 | |
| *** jcoufal has joined #openstack-ironic | 21:29 | |
| *** jcoufal_ has joined #openstack-ironic | 21:34 | |
| *** jcoufal has quit IRC | 21:35 | |
| *** jcoufal has joined #openstack-ironic | 21:37 | |
| *** ribe has joined #openstack-ironic | 21:37 | |
| openstackgerrit | Lin Yang proposed openstack/sushy master: Fix wrong message of invalid boot "enabled" parameter https://review.openstack.org/549503 | 21:38 |
| *** jcoufal_ has quit IRC | 21:39 | |
| *** jcoufal has quit IRC | 21:41 | |
| *** jcoufal has joined #openstack-ironic | 21:43 | |
| *** gyee has quit IRC | 21:50 | |
| *** bfournie has quit IRC | 21:50 | |
| *** baha has quit IRC | 21:50 | |
| *** jcoufal has quit IRC | 22:01 | |
| *** e0ne has quit IRC | 22:12 | |
| *** AlexeyAbashkin has joined #openstack-ironic | 22:20 | |
| openstackgerrit | Merged openstack/ironic-inspector master: Switch documentation to hardware types https://review.openstack.org/545285 | 22:22 |
| *** AlexeyAbashkin has quit IRC | 22:25 | |
| *** rcernin has joined #openstack-ironic | 22:34 | |
| *** Goneri has quit IRC | 22:37 | |
| *** milan has quit IRC | 22:41 | |
| *** bfournie has joined #openstack-ironic | 23:07 | |
| *** rpioso is now known as rpioso|afk | 23:07 | |
| *** cdearborn has quit IRC | 23:18 | |
| *** AlexeyAbashkin has joined #openstack-ironic | 23:20 | |
| *** AlexeyAbashkin has quit IRC | 23:24 | |
| *** masber has joined #openstack-ironic | 23:27 | |
| *** rbudden has quit IRC | 23:40 | |
| *** Sukhdev_ has quit IRC | 23:44 | |
| *** rama_y has quit IRC | 23:46 | |
| *** r-daneel has quit IRC | 23:49 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!