| *** smoriya has joined #openstack-ironic | 00:02 | |
| *** yuanying has joined #openstack-ironic | 00:03 | |
| *** naohirot has joined #openstack-ironic | 00:05 | |
| *** spandhe has joined #openstack-ironic | 00:06 | |
| *** Sukhdev has quit IRC | 00:07 | |
| *** Sukhdev has joined #openstack-ironic | 00:07 | |
| jlvillal | mrda, Woo hoo on your spec :) | 00:15 |
|---|---|---|
| mrda | thanks jlvillal | 00:17 |
| mrda | next is the code :) | 00:17 |
| jlvillal | mrda, +1 :) | 00:17 |
| mrda | and such it's such a small change i'm hoping it won't be controversial | 00:18 |
| *** edcragg has quit IRC | 00:27 | |
| *** korean101 has quit IRC | 00:28 | |
| *** spandhe has quit IRC | 00:29 | |
| *** amotoki has joined #openstack-ironic | 00:33 | |
| *** spandhe has joined #openstack-ironic | 00:34 | |
| *** alexpilotti has quit IRC | 00:36 | |
| *** amotoki has quit IRC | 00:38 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic: Updated from global requirements https://review.openstack.org/190357 | 00:42 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic-lib: Updated from global requirements https://review.openstack.org/189900 | 00:42 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic-python-agent: Updated from global requirements https://review.openstack.org/190427 | 00:42 |
| *** mtanino has quit IRC | 00:44 | |
| Sukhdev | devananda: Ping | 00:45 |
| *** saripurigopi has joined #openstack-ironic | 00:47 | |
| *** ijw has quit IRC | 00:50 | |
| *** ijw has joined #openstack-ironic | 00:50 | |
| *** spandhe has quit IRC | 00:55 | |
| *** chlong has quit IRC | 01:11 | |
| *** pshige has quit IRC | 01:18 | |
| *** kan_ has joined #openstack-ironic | 01:37 | |
| *** yog__ has joined #openstack-ironic | 01:54 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Add iRMC vendor passthru for soft reboot and NMI https://review.openstack.org/187082 | 02:02 |
| *** Sukhdev has quit IRC | 02:07 | |
| *** chenglch has joined #openstack-ironic | 02:15 | |
| *** saripurigopi has quit IRC | 02:15 | |
| *** rloo has quit IRC | 02:16 | |
| openstackgerrit | Merged openstack/ironic: Add pxe_ucs and agent_ucs drivers to manage Cisco UCS servers https://review.openstack.org/159734 | 02:17 |
| *** amotoki has joined #openstack-ironic | 02:21 | |
| *** kan_ has quit IRC | 02:22 | |
| *** saripurigopi has joined #openstack-ironic | 02:22 | |
| *** saripurigopi has quit IRC | 02:24 | |
| *** saripurigopi has joined #openstack-ironic | 02:31 | |
| *** saripurigopi has quit IRC | 02:36 | |
| *** kan_ has joined #openstack-ironic | 02:37 | |
| *** amotoki has quit IRC | 02:40 | |
| *** ramineni has joined #openstack-ironic | 02:43 | |
| *** ijw has quit IRC | 02:50 | |
| *** yuanying_ has joined #openstack-ironic | 02:58 | |
| *** yuanying has quit IRC | 03:01 | |
| *** david-lyle has joined #openstack-ironic | 03:04 | |
| *** yuanying has joined #openstack-ironic | 03:20 | |
| *** yuanyin__ has joined #openstack-ironic | 03:20 | |
| *** yuanyin__ has quit IRC | 03:21 | |
| *** yuanyin__ has joined #openstack-ironic | 03:22 | |
| *** yuanying_ has quit IRC | 03:22 | |
| *** yuanying has quit IRC | 03:24 | |
| *** yuanyin__ has quit IRC | 03:25 | |
| *** saripurigopi has joined #openstack-ironic | 03:26 | |
| *** ijw has joined #openstack-ironic | 03:28 | |
| *** yuanying has joined #openstack-ironic | 03:34 | |
| *** ijw_ has joined #openstack-ironic | 03:35 | |
| *** rameshg87 has joined #openstack-ironic | 03:36 | |
| rameshg87 | good morning ironic | 03:37 |
| Haomeng|2 | rameshg87: morning ramesh:) | 03:37 |
| rameshg87 | morning Haomeng|2 | 03:37 |
| Haomeng|2 | rameshg87: :) | 03:38 |
| *** wanghua has joined #openstack-ironic | 03:38 | |
| *** yuanying has quit IRC | 03:38 | |
| *** ijw has quit IRC | 03:39 | |
| *** yuanying has joined #openstack-ironic | 03:41 | |
| *** yuanying has quit IRC | 03:42 | |
| *** yuanying has joined #openstack-ironic | 03:43 | |
| *** wanghua has quit IRC | 03:46 | |
| *** wanghua has joined #openstack-ironic | 03:47 | |
| *** yuanying has quit IRC | 03:48 | |
| *** amotoki has joined #openstack-ironic | 03:49 | |
| saripurigopi | Morning rameshg87, Haomeng|2 | 03:50 |
| rameshg87 | morning saripurigopi | 03:53 |
| *** yuanying has joined #openstack-ironic | 03:55 | |
| saripurigopi | rameshg87: reg https://review.openstack.org/#/c/159734/, for the missing testcases, should I submit a bug ? | 03:57 |
| *** ijw_ has quit IRC | 03:57 | |
| rameshg87 | saripurigopi: no need, you can just add a follow up patch | 03:58 |
| saripurigopi | rameshg87: sure, will do that. | 03:58 |
| *** yuanying has quit IRC | 03:58 | |
| rameshg87 | saripurigopi: something like https://review.openstack.org/#/c/187091/ | 03:59 |
| rameshg87 | saripurigopi: you can mention the change id of the other review in the commit msg so that it is easy to track | 03:59 |
| saripurigopi | rameshg87: okay | 03:59 |
| *** zz_natorious is now known as natorious | 04:02 | |
| *** yuanying has joined #openstack-ironic | 04:05 | |
| rameshg87 | kan_: hi | 04:24 |
| openstackgerrit | Michael Davies proposed openstack/python-ironicclient: Cache negotiated api microversion for server https://review.openstack.org/173674 | 04:25 |
| *** natorious is now known as zz_natorious | 04:38 | |
| kan_ | rameshg87: hi :) | 04:55 |
| *** achanda has joined #openstack-ironic | 04:59 | |
| rameshg87 | kan_: hi | 05:02 |
| rameshg87 | kan_: I was just checking https://review.openstack.org/#/c/181809/3 | 05:02 |
| *** ijw has joined #openstack-ironic | 05:03 | |
| rameshg87 | kan_: I just hit the issue because my oslo.db was old | 05:03 |
| kan_ | rameshg87: there is a oslo.db patch that need to pass the test: https://review.openstack.org/#/c/180206/ | 05:05 |
| rameshg87 | kan_: yeah | 05:05 |
| kan_ | rameshg87: without that patch, the test case can not pass | 05:05 |
| rameshg87 | kan_: it seems 1.11.0 has that fix | 05:06 |
| rameshg87 | kan_: any idea if 1.10.0 has that ? it seems that's what is recommended by ironic now | 05:06 |
| rameshg87 | kan_: yeah 1.10.0 is good | 05:08 |
| rameshg87 | just confirmed | 05:08 |
| kan_ | rameshg87: yes, the requirements in ironic is 1.10.0 now, checked that, the 1.10.0 contanis the oslo.db patch now | 05:10 |
| rameshg87 | kan_: yeah, thanks | 05:10 |
| kan_ | rameshg87: wlc :) | 05:10 |
| *** lazy_prince has joined #openstack-ironic | 05:11 | |
| lazy_prince | morning everyone.. o/ | 05:11 |
| *** kan_ has left #openstack-ironic | 05:12 | |
| rameshg87 | lazy_prince: o/ | 05:14 |
| *** kan_ has joined #openstack-ironic | 05:14 | |
| *** achanda has quit IRC | 05:22 | |
| Haomeng|2 | saripurigopi: morning:) | 05:33 |
| *** Haomeng|2 is now known as Haomeng | 05:33 | |
| *** Nisha has joined #openstack-ironic | 05:40 | |
| *** kalpase has joined #openstack-ironic | 05:43 | |
| *** coolsvap|afk is now known as coolsvap | 05:50 | |
| *** lsmola has joined #openstack-ironic | 05:57 | |
| *** ukalifon has joined #openstack-ironic | 05:58 | |
| *** kan_ has quit IRC | 05:58 | |
| *** kan_ has joined #openstack-ironic | 05:59 | |
| *** achanda has joined #openstack-ironic | 05:59 | |
| *** mdbooth has quit IRC | 06:00 | |
| *** mdbooth has joined #openstack-ironic | 06:02 | |
| *** heyongli has quit IRC | 06:06 | |
| *** heyongli- is now known as heyongli | 06:06 | |
| *** yonglihe has joined #openstack-ironic | 06:07 | |
| *** dtantsur|afk is now known as dtantsur | 06:22 | |
| dtantsur | Morning! | 06:22 |
| *** absubram has quit IRC | 06:28 | |
| *** achanda has quit IRC | 06:32 | |
| *** ijw has quit IRC | 06:33 | |
| saripurigopi | morning dtantsur | 06:35 |
| dtantsur | o/ | 06:35 |
| *** absubram has joined #openstack-ironic | 06:39 | |
| *** achanda has joined #openstack-ironic | 06:41 | |
| *** kalpase has quit IRC | 06:45 | |
| *** kalpase has joined #openstack-ironic | 06:49 | |
| *** jcoufal has joined #openstack-ironic | 06:49 | |
| *** kalpase has quit IRC | 06:51 | |
| *** ifarkas has joined #openstack-ironic | 06:52 | |
| *** kalpase has joined #openstack-ironic | 06:52 | |
| rameshg87 | dtantsur: o/ | 07:03 |
| dtantsur | o/ | 07:03 |
| *** Haomeng has quit IRC | 07:06 | |
| *** kan_ has quit IRC | 07:08 | |
| *** kan_ has joined #openstack-ironic | 07:08 | |
| *** e0ne has joined #openstack-ironic | 07:08 | |
| openstackgerrit | Pratyusha proposed stackforge/proliantutils: Add iSCSI boot support in RIS https://review.openstack.org/187162 | 07:11 |
| *** e0ne has quit IRC | 07:12 | |
| openstackgerrit | Merged stackforge/proliantutils: Activate iLO license https://review.openstack.org/190030 | 07:14 |
| *** subscope has joined #openstack-ironic | 07:15 | |
| *** jrist has quit IRC | 07:23 | |
| *** yuikotakada has joined #openstack-ironic | 07:25 | |
| *** ndipanov has quit IRC | 07:26 | |
| *** bradjones has quit IRC | 07:27 | |
| *** e0ne has joined #openstack-ironic | 07:28 | |
| *** bradjones has joined #openstack-ironic | 07:29 | |
| *** bradjones has quit IRC | 07:29 | |
| *** bradjones has joined #openstack-ironic | 07:29 | |
| *** e0ne has quit IRC | 07:29 | |
| *** achanda has quit IRC | 07:33 | |
| *** jistr has joined #openstack-ironic | 07:33 | |
| *** stendulker has joined #openstack-ironic | 07:54 | |
| *** achanda has joined #openstack-ironic | 07:57 | |
| *** achanda_ has joined #openstack-ironic | 08:08 | |
| *** achanda has quit IRC | 08:11 | |
| *** MattMan has quit IRC | 08:11 | |
| *** achanda_ has quit IRC | 08:13 | |
| *** romainh has joined #openstack-ironic | 08:18 | |
| *** lucasagomes has joined #openstack-ironic | 08:20 | |
| *** athomas has joined #openstack-ironic | 08:24 | |
| *** jistr has quit IRC | 08:27 | |
| *** oomichi has joined #openstack-ironic | 08:33 | |
| openstackgerrit | Merged openstack/ironic: Updated from global requirements https://review.openstack.org/190357 | 08:35 |
| openstackgerrit | Merged openstack/ironic: Install guide reflects changes on master branch https://review.openstack.org/189327 | 08:36 |
| *** erelwell has joined #openstack-ironic | 08:37 | |
| *** erelwell is now known as bethelwell | 08:38 | |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add vendor-passthru to attach and boot an ISO https://review.openstack.org/188317 | 08:39 |
| *** MattMan has joined #openstack-ironic | 08:41 | |
| *** jistr has joined #openstack-ironic | 08:43 | |
| rameshg87 | lucasagomes: hi | 08:44 |
| lucasagomes | rameshg87, hello there | 08:44 |
| rameshg87 | lucasagomes: have you started working on https://review.openstack.org/#/c/185667/ ? | 08:44 |
| rameshg87 | lucasagomes: if not I am going to get to it now | 08:44 |
| lucasagomes | rameshg87, go for it | 08:45 |
| lucasagomes | I've tested the current code | 08:45 |
| lucasagomes | and it works for me | 08:45 |
| rameshg87 | great | 08:45 |
| * rameshg87 starts fixing it | 08:45 | |
| lucasagomes | but I haven't modified it yet to add the new options | 08:45 |
| rameshg87 | okay, I am doing them right away | 08:45 |
| lucasagomes | ack | 08:45 |
| lucasagomes | thanks | 08:45 |
| *** ndipanov has joined #openstack-ironic | 08:46 | |
| *** kalpase has quit IRC | 08:50 | |
| openstackgerrit | IWAMOTO Toshihiro proposed openstack/ironic-specs: Collect IPA logs https://review.openstack.org/168799 | 08:55 |
| *** chenglch|2 has joined #openstack-ironic | 09:05 | |
| *** enikanorov_ has quit IRC | 09:06 | |
| *** chenglch has quit IRC | 09:06 | |
| *** enikanorov has joined #openstack-ironic | 09:06 | |
| *** dguerri` is now known as dguerri | 09:08 | |
| *** pelix has joined #openstack-ironic | 09:24 | |
| *** edcragg has joined #openstack-ironic | 09:31 | |
| *** pelix has quit IRC | 09:32 | |
| *** Nisha has quit IRC | 09:33 | |
| *** Nisha_away has joined #openstack-ironic | 09:33 | |
| *** oomichi has quit IRC | 09:39 | |
| *** kan_ has quit IRC | 09:40 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic-specs: Enhance Power Interface for Soft Reboot and NMI https://review.openstack.org/186700 | 09:40 |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic: IPA: Do a soft power off at the end of deployment https://review.openstack.org/185667 | 09:41 |
| *** divius has joined #openstack-ironic | 09:41 | |
| * rameshg87 goes home | 09:43 | |
| *** rameshg87 has quit IRC | 09:43 | |
| *** achanda has joined #openstack-ironic | 09:44 | |
| *** e0ne has joined #openstack-ironic | 09:47 | |
| *** divius has left #openstack-ironic | 09:48 | |
| *** e0ne is now known as e0ne_ | 09:54 | |
| *** naohirot has quit IRC | 09:54 | |
| *** Nisha_away has quit IRC | 09:58 | |
| *** e0ne_ has quit IRC | 10:00 | |
| *** e0ne has joined #openstack-ironic | 10:01 | |
| openstackgerrit | Pratyusha proposed stackforge/proliantutils: Add iSCSI boot support in RIS https://review.openstack.org/187162 | 10:03 |
| *** achanda has quit IRC | 10:04 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Wake-On-Lan Power interface https://review.openstack.org/179078 | 10:05 |
| *** pelix has joined #openstack-ironic | 10:06 | |
| *** chenglch|2 has quit IRC | 10:09 | |
| *** zhenguo has quit IRC | 10:09 | |
| *** achanda has joined #openstack-ironic | 10:10 | |
| openstackgerrit | Pratyusha proposed stackforge/proliantutils: Add iSCSI boot support in RIS https://review.openstack.org/187162 | 10:11 |
| *** achanda has quit IRC | 10:15 | |
| openstackgerrit | Merged openstack/ironic: Sync with latest oslo-incubator https://review.openstack.org/189110 | 10:15 |
| *** yuikotakada has quit IRC | 10:15 | |
| *** yuanying has quit IRC | 10:18 | |
| *** achanda has joined #openstack-ironic | 10:21 | |
| *** achanda has quit IRC | 10:22 | |
| *** athomas has quit IRC | 10:23 | |
| *** athomas has joined #openstack-ironic | 10:23 | |
| *** yuanying has joined #openstack-ironic | 10:26 | |
| *** sambetts_ has joined #openstack-ironic | 10:27 | |
| sambetts_ | Good UGT morning ironic :) | 10:28 |
| openstackgerrit | Pratyusha proposed stackforge/proliantutils: Add iSCSI boot support in RIS https://review.openstack.org/187162 | 10:29 |
| sambetts_ | dtantsur: Have you had a chance to look at the latest version of the not found hook patch? I'm getting a pep8 error I'm not sure how to fix without refactoring the process function | 10:31 |
| dtantsur | sambetts, morning, will have a look | 10:32 |
| *** e0ne is now known as e0ne_ | 10:33 | |
| *** e0ne_ is now known as e0ne | 10:34 | |
| *** jrist has joined #openstack-ironic | 10:34 | |
| dtantsur | sambetts_, oh cool :) yeah, pep8 is right, the function is way too big now. maybe we should split away the whole node lookup part? aka def fine_node(introspection_data) -> node_info? | 10:36 |
| *** stendulker has quit IRC | 10:36 | |
| *** romainh has quit IRC | 10:36 | |
| sambetts_ | dtantsur: Yeah, that might be required, I was trying to pull out just the bit I was adding, but because of the whole failures being collected in a list thing, I just found I was passing that list around all over the place | 10:41 |
| *** yog__ has quit IRC | 10:45 | |
| *** ramineni has quit IRC | 10:55 | |
| *** athomas has quit IRC | 10:55 | |
| *** romainh has joined #openstack-ironic | 10:56 | |
| sambetts_ | dtantsur: I've never seen flake8 check for complexity before, does it always do that or is that something you've enabled for inspector? | 10:56 |
| dtantsur | sambetts_, it's enabled in tox.ini | 10:57 |
| sambetts_ | dtantsur: ah, thats why I've not seen it in other projects then, they must have it disabled | 10:57 |
| dtantsur | sambetts_, yeah, not sure if other projects have it | 10:57 |
| sambetts_ | dtantsur: I've definatly worked on functions more complex than process is before and never seen it :-P | 10:58 |
| dtantsur | :) | 10:58 |
| *** kbyrne has joined #openstack-ironic | 11:03 | |
| *** achanda has joined #openstack-ironic | 11:07 | |
| *** yuanying has quit IRC | 11:08 | |
| *** athomas has joined #openstack-ironic | 11:11 | |
| *** subscope has quit IRC | 11:12 | |
| *** openstackgerrit has quit IRC | 11:13 | |
| *** openstackgerrit has joined #openstack-ironic | 11:14 | |
| *** subscope has joined #openstack-ironic | 11:14 | |
| *** e0ne is now known as e0ne_ | 11:21 | |
| *** achanda has quit IRC | 11:25 | |
| *** dlpartain has joined #openstack-ironic | 11:25 | |
| *** dlpartain has left #openstack-ironic | 11:25 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Wake-On-Lan Power interface https://review.openstack.org/179078 | 11:31 |
| *** e0ne_ has quit IRC | 11:32 | |
| *** e0ne has joined #openstack-ironic | 11:36 | |
| *** derekh has joined #openstack-ironic | 11:49 | |
| *** saripurigopi has quit IRC | 11:53 | |
| TheJulia | good morning everyone | 11:54 |
| *** jrist has quit IRC | 11:55 | |
| sambetts_ | morning TheJulia | 11:55 |
| *** jrist has joined #openstack-ironic | 11:59 | |
| *** yuikotakada has joined #openstack-ironic | 12:00 | |
| dtantsur | TheJulia, morning | 12:00 |
| *** alexpilotti has joined #openstack-ironic | 12:01 | |
| *** thrash|g0ne is now known as thrash | 12:06 | |
| *** yuikotakada has quit IRC | 12:06 | |
| *** yuikotakada has joined #openstack-ironic | 12:06 | |
| *** trown|outttypeww is now known as trown | 12:07 | |
| *** alexpilotti has quit IRC | 12:09 | |
| *** afaranha has joined #openstack-ironic | 12:13 | |
| *** absubram has quit IRC | 12:38 | |
| *** jjohnson2 has joined #openstack-ironic | 12:38 | |
| *** dprince has joined #openstack-ironic | 12:39 | |
| *** derekh has quit IRC | 12:40 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic-specs: Add pluggable credentials storage https://review.openstack.org/186056 | 12:42 |
| *** derekh has joined #openstack-ironic | 12:42 | |
| *** smoriya has quit IRC | 12:45 | |
| *** wanghua has quit IRC | 12:48 | |
| *** amotoki has quit IRC | 12:50 | |
| *** jcoufal has quit IRC | 12:51 | |
| *** rloo has joined #openstack-ironic | 13:03 | |
| *** trown is now known as trown|outttypeww | 13:05 | |
| dtantsur | ifarkas, hi, could you please review https://review.openstack.org/#/c/189346/ ? | 13:10 |
| ifarkas | dtantsur, sure | 13:11 |
| *** cdearborn has joined #openstack-ironic | 13:14 | |
| *** sambetts_ has quit IRC | 13:20 | |
| *** yuikotakada has quit IRC | 13:26 | |
| *** alexpilotti has joined #openstack-ironic | 13:46 | |
| *** yuikotakada has joined #openstack-ironic | 13:47 | |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Rework processing hook interface for 2.0.0 https://review.openstack.org/189346 | 13:48 |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Move create_ports to NodeInfo https://review.openstack.org/190643 | 13:48 |
| jroll | I'd like to hear folks' thoughts about enabling agent log shipping by default. I disagree, but I can live with it. https://review.openstack.org/#/c/168799/ | 14:00 |
| jroll | also good morning :) | 14:00 |
| *** bethelwe_ has joined #openstack-ironic | 14:00 | |
| jlvillal | Good morning Ironic | 14:02 |
| *** bethelwell has quit IRC | 14:03 | |
| *** achanda has joined #openstack-ironic | 14:03 | |
| openstackgerrit | Merged stackforge/ironic-discoverd: Rework processing hook interface for 2.0.0 https://review.openstack.org/189346 | 14:05 |
| dtantsur | jroll, I understood that it's logged as INFO by default, right? so it won't be visible unless you set verbose=true? | 14:06 |
| dtantsur | and morning jroll, jlvillal | 14:06 |
| jlvillal | sambetts, dtantsur: Complexity checking is enabled in openstack/ironic. So discoverd is not the only project :) | 14:07 |
| *** achanda has quit IRC | 14:07 | |
| dtantsur | ah good :) | 14:07 |
| dtantsur | well, I suspected I've copied it actually from some project :D | 14:08 |
| *** edcragg has quit IRC | 14:09 | |
| jroll | dtantsur: yeah, that's why I might be ok with it, though I tend to think most people run openstack in DEBUG in general | 14:09 |
| jroll | I think someone's going to get a surprise when they turn debug on and everything grinds to a halt | 14:09 |
| dtantsur | well, we do as well actually... | 14:10 |
| jroll | or when they upgrade their current environment and don't disable that | 14:10 |
| dtantsur | but I suppose people enabling DEBUG should generally understand they're going to have a lot of logs... | 14:10 |
| * dtantsur is not decided as well | 14:10 | |
| jroll | it's not about the amount of logs, it's about sending and unpacking a json object with megabytes of logs :) | 14:10 |
| * dtantsur reads the spec again | 14:11 | |
| jroll | I -1'd for now but I'd like to talk with folks about it today | 14:11 |
| jroll | dtantsur: also +2 on enroll spec | 14:11 |
| *** achanda has joined #openstack-ironic | 14:11 | |
| dtantsur | \o/ | 14:12 |
| dtantsur | who is gonna help me land the enroll spec now? lucasagomes? ;) | 14:12 |
| jroll | heh | 14:13 |
| lucasagomes | yo morning... oh I have to re-read it :-( | 14:13 |
| lucasagomes | I'm fighting neutron internally | 14:13 |
| lucasagomes | mind worth asking you guys, what I'm seem is http://paste.openstack.org/show/284017/ | 14:14 |
| lucasagomes | I've changed sudoers already... checked permissions | 14:14 |
| lucasagomes | kinda running out of ideas now | 14:14 |
| NobodyCam | good morning Ironicers | 14:15 |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: When boot option is not persisted, set boot on next power on https://review.openstack.org/177642 | 14:15 |
| jroll | lucasagomes: can that user get a shell? | 14:15 |
| jroll | morning NobodyCam jlvillal lucasagomes :) | 14:16 |
| dtantsur | NobodyCam, morning! | 14:16 |
| dtantsur | NobodyCam, can I grab you to have a look at the enroll spec https://review.openstack.org/#/c/179151 please? | 14:16 |
| NobodyCam | morning jroll dtantsur and lucasagomes | 14:16 |
| *** achanda has quit IRC | 14:17 | |
| *** achanda has joined #openstack-ironic | 14:17 | |
| dtantsur | jroll, re "Many people run OpenStack at DEBUG level (because it's impossible to find out what went wrong otherwise)", I think it's our bug actually :) INFO level should give a good idea already, DEBUG should be only for hard issues, where you need to know really EVERYTHING | 14:17 |
| lucasagomes | jroll, nop passwd is pointing to /sbin/nologin | 14:18 |
| lucasagomes | but other openstack users are the same | 14:18 |
| BadCub | morning folks | 14:18 |
| lucasagomes | NobodyCam, morning | 14:18 |
| NobodyCam | :) | 14:18 |
| NobodyCam | mornign BadCub | 14:18 |
| jroll | lucasagomes: hrm, that's my only idea | 14:19 |
| dtantsur | BadCub, morning | 14:19 |
| jroll | dtantsur: sure, but what happens when one build failed in a strange way and you don't have the data you need to debug it | 14:19 |
| BadCub | morning NobodyCam dtantsur jroll lucasagomes | 14:19 |
| jroll | dtantsur: fwiw I think no matter how good INFO is I'll always run at DEBUG just in case | 14:20 |
| jroll | hiya BadCub :) | 14:20 |
| dtantsur | jroll, then you'll probably need IPA logs too :) | 14:20 |
| lucasagomes | jroll, :-( | 14:20 |
| lucasagomes | and selinux is permissive so... it's not even selinux | 14:20 |
| lucasagomes | gosh | 14:20 |
| jroll | dtantsur: meh, in my experience IPA doesn't fail in strange ways, we try to keep it dumb for a reason | 14:20 |
| jroll | dtantsur: I'd like IPA logs but I don't think pushing them through our API is the way to do it :) | 14:21 |
| dtantsur | yeah, some specific log solution sounds better to me... | 14:21 |
| rloo | hello and good morning ironic'ers | 14:21 |
| dtantsur | rloo, morning! | 14:22 |
| NobodyCam | morning rloo :) | 14:22 |
| rloo | wrt microversioning, are you all in agreement that if a new vendor-passthru method is added, no bump is needed? | 14:22 |
| jroll | morning rloo :) | 14:22 |
| *** achanda has quit IRC | 14:22 | |
| dtantsur | jroll, like we discussed with lucasagomes that maybe we should not have used API for callbacks (e.g. ramdisks). we discussed using AMQP instead | 14:22 |
| NobodyCam | uggh back to back conf calls this morning :-p | 14:22 |
| jroll | dtantsur: ahhhhhhhhhhhhhhhhhhh | 14:22 |
| dtantsur | rloo, I am, people are not. | 14:22 |
| * jroll runs away | 14:22 | |
| lucasagomes | RPC notifications | 14:22 |
| dtantsur | lol | 14:23 |
| rloo | in reference to https://review.openstack.org/#/c/188317/ | 14:23 |
| jroll | dtantsur: thousands of connections to a single rabbit seems not fun | 14:23 |
| lucasagomes | rloo, yeah, we can't guarantee anything with vendor stuff right? | 14:23 |
| rloo | dtantsur: i can also send email about it :-) | 14:23 |
| lucasagomes | backwards compat or anything | 14:23 |
| lucasagomes | so no bump | 14:23 |
| *** yuikotakada has quit IRC | 14:23 | |
| rloo | lucasagomes: oh, so we don't guarantee backwards compat on passthru methods? | 14:23 |
| dtantsur | rloo, lucasagomes, people demand mversion bump for every API change, period | 14:23 |
| dtantsur | (and I think it's not sane, but who cares?) | 14:24 |
| rloo | dtantsur: I'm just a follower. baaaa... | 14:24 |
| lucasagomes | we can't do for vendors, we have no contract on that api | 14:24 |
| lucasagomes | rloo, no | 14:24 |
| dtantsur | that's one of my objections to the guidelines | 14:25 |
| dtantsur | vendor passthru ARE API, but we can't seriously version them for a lot of reasons | 14:25 |
| dtantsur | my harder objection is versioning of indirect API modifications, like new fields in driver_info | 14:25 |
| lucasagomes | https://www.mail-archive.com/openstack-dev@lists.openstack.org/msg54752.html | 14:25 |
| dtantsur | was it accepted into the guideline? | 14:26 |
| lucasagomes | see the next ans by devananda | 14:26 |
| lucasagomes | yeah I mean, the way I see it is like the guy described... that vendor endpoint is open group | 14:26 |
| lucasagomes | open ground* | 14:26 |
| jroll | dtantsur: god, I hope not, we can't seriously version a json field by api version | 14:27 |
| dtantsur | then we can't version "all user visible changes" even remotely :) | 14:27 |
| jroll | I mean... maybe driver_info should be versioned :P | 14:27 |
| jroll | {'version': 2, ...} | 14:27 |
| dtantsur | ouch | 14:27 |
| jroll | but more seriously, code can write there at will etc | 14:28 |
| *** zz_jgrimm is now known as jgrimm | 14:28 | |
| dtantsur | next crazy question: we have only partial versioning for new states | 14:28 |
| jroll | I think anything an out of tree driver can change shouldn't be versioned | 14:28 |
| rloo | lucasagomes: if i use deva's description about users wanting to move workload between openstack clouds, seems like the passthru methods is same. one cloud can handle it, the other cloud barfs. | 14:28 |
| jroll | and so that includes json fields and vendor passthru | 14:28 |
| lucasagomes | rloo, right, but that's addons | 14:28 |
| lucasagomes | the standard (core api) should work | 14:28 |
| lucasagomes | vendor is to enhance the user xperience, it's not required to work | 14:28 |
| dtantsur | rloo, what was said in this thread is all correct and great, but also idealistic | 14:29 |
| rloo | lucasagomes: weren't you the one that said that heartbeat and lookup would continue to be vendor passthru? | 14:29 |
| dtantsur | I don't believe people won't add downstream drivers and vendor passthru | 14:29 |
| rloo | lucasagomes: seems like those are core methods. or are going to be... | 14:29 |
| lucasagomes | rloo, I'm ok bumping it | 14:29 |
| rloo | i hate this microversioning. devil is in the details. | 14:29 |
| lucasagomes | yeah :-/ it's really complicated | 14:29 |
| rloo | lucasagomes: i'm ok not bumping it. but i think we need to understand and agree on something. | 14:29 |
| jroll | dtantsur: right. I think anything an out of tree driver can change shouldn't be versioned. that includes json fields and vendor passthru. | 14:29 |
| jroll | rloo: it's almost like we need solid docs on when to bump and when not to bump | 14:30 |
| lucasagomes | I don't it just requires a lot of thinking before doing any micro version stuff... If we try to come up with a hard rule when to bump it seems that we won't be able to | 14:30 |
| dtantsur | jroll, then we have have-baked versioning? they call it "contract" even :D | 14:30 |
| jroll | rloo: I feel like that's why people don't like it, they are unsure? | 14:30 |
| rloo | jroll: I WANT solid doc! | 14:30 |
| jroll | dtantsur: we just explicitly say there's no contract for those. | 14:30 |
| rloo | jroll: I think some people don't like it cuz they don't agree with the idea. Me, I don't care for it but I don't care to argue against it cuz it has some merit, so me, yes, i want doc so everyone is on the same wavelength.Otherwise it becomes a mess. | 14:31 |
| dtantsur | jroll, but that's the goal of microversioning guidelines as I was told in that ML | 14:31 |
| jroll | dtantsur: e.g. nova doesn't have a contract for keys in instance metadata | 14:31 |
| *** absubram has joined #openstack-ironic | 14:31 | |
| * lucasagomes also thinks microversioning is a mess, 3 headed unicorn | 14:33 | |
| dtantsur | very ++ not wow at all | 14:34 |
| jroll | I think it's been painful reviewing | 14:34 |
| jroll | as a user I've been loving it | 14:34 |
| lucasagomes | maybe that's becuase the design of our api doesn't suit the idea of microversining | 14:34 |
| *** e0ne is now known as e0ne_ | 14:34 | |
| lucasagomes | we chose json fields because of the flexibility and all | 14:35 |
| lucasagomes | same for vendor passthru | 14:35 |
| lucasagomes | but flexibilitty doesn't play well with microversioning | 14:35 |
| lucasagomes | and we need flexibility because at the beggining we had no idea how the API would look like for something like Ironic | 14:35 |
| NobodyCam | lucasagomes: lol cerberus the three headed (pink) unicorn | 14:35 |
| lucasagomes | (maybe even now we don't) | 14:35 |
| * dtantsur definitely does not | 14:37 | |
| jroll | I want to reiterate that having microversions has been really awesome from an ops/user perspective | 14:38 |
| dtantsur | jroll, for which use case? for strict contract or for not breaking compatibility? | 14:39 |
| lucasagomes | jroll, cause of the NOSTATE -> AVAILABLE change? | 14:39 |
| lucasagomes | I mean, it's already in. So now we gotta try to make the best use of it | 14:40 |
| rloo | I suspect we'd all agree that the microversioning is good for non-backwards-compatible changes. | 14:40 |
| lucasagomes | complaining doesn't solve the problem | 14:40 |
| *** achanda has joined #openstack-ironic | 14:40 | |
| dtantsur | rloo, ++ | 14:40 |
| dtantsur | though the guideline IIRC requires us to bump non-micro version in this case | 14:40 |
| rloo | we just need to know when to bump for 'new' changes | 14:40 |
| rloo | dtantsur: oh. didn't know that. on my list to read the guidelines sigh but was hoping others here would be the experts. | 14:41 |
| dtantsur | my memory may let me down though, but that's how I understood: | 14:41 |
| dtantsur | microversion bump for every change, macroversion bump for breaking ones | 14:41 |
| *** lazy_prince has quit IRC | 14:41 | |
| dtantsur | like semver | 14:41 |
| jroll | dtantsur: lucasagomes: primarily for not breaking things. the strict contract thing could be useful, but if we're being real nobody is ever going to be a user of two ironic systems | 14:41 |
| dtantsur | jroll, that's my point as well: it's great for not breaking people (like with AVAILABLE or future ENROLL) | 14:42 |
| devananda | jroll: there will totally be people who write a thing that should work on >1 ironic endpoint | 14:42 |
| devananda | jroll: because $regions | 14:42 |
| dtantsur | how did they manage doing it before? ;) | 14:42 |
| *** lazy_prince has joined #openstack-ironic | 14:43 | |
| jroll | devananda: but will those regions be different enough to matter? | 14:43 |
| dtantsur | I mean, I stay sure that microversions won't make life easy for many people, unless used only for breaking changes | 14:43 |
| devananda | jroll: they had better not be -- but in actuality, they could be | 14:43 |
| devananda | jroll: as it, the control plane services may well be managed by different people than the software deployed by ironic | 14:43 |
| devananda | s/it/in/ | 14:44 |
| jroll | devananda: urgh, yeah, didn't have enough coffee to consider the non-nova use case iow real users hitting ironic directly | 14:44 |
| *** achanda has quit IRC | 14:45 | |
| * jroll wonders if one day there will be a public cloud where users talk to ironic | 14:45 | |
| lucasagomes | heh | 14:45 |
| *** e0ne_ has quit IRC | 14:45 | |
| NobodyCam | thats just an evil thought jroll | 14:46 |
| jroll | why? | 14:46 |
| jroll | I'm totally serious | 14:46 |
| NobodyCam | we have nothing in place for it | 14:47 |
| jroll | we're only really lacking the concept of tenants and policy | 14:47 |
| jroll | *today* we don't :) | 14:47 |
| NobodyCam | :) | 14:47 |
| jroll | to be clear, I never want to run that cloud, but yeah. it could happen. | 14:47 |
| * dtantsur wonders how many digits will our microversion have by that day | 14:49 | |
| rloo | dtantsur: not that many cuz of the macroversion bumps :) | 14:50 |
| dtantsur | right | 14:50 |
| jroll | urgh, why bump macroversions? | 14:50 |
| jroll | that's like super painful no? | 14:50 |
| devananda | jroll: i still see value in nova's API as a common abstraction for vm/bm/container | 14:50 |
| *** jcoufal has joined #openstack-ironic | 14:50 | |
| rloo | devananda: if you were lurking/scrolling back, what say you wrt microversion bump for new vendor-passthru method? | 14:50 |
| dtantsur | "dear users, today we release Ironic Z, please update your scripts to API version 19.23" | 14:50 |
| devananda | jroll: macroversion if we ever completely change the semantics. which we should never do, but it might happen anyway | 14:51 |
| jroll | devananda: I'm not saying nova isn't valueable, I'm just wondering out loud :) | 14:51 |
| devananda | lemme read back that far | 14:51 |
| devananda | jroll: ah :) | 14:51 |
| jroll | devananda: sure, folks are talking about macroversion bump for any breaking changes apparently | 14:51 |
| lucasagomes | would be cool if nova was just like Ironic an API for VMs | 14:51 |
| lucasagomes | and another project would take care of flavors, instances etc | 14:51 |
| jroll | lucasagomes: it's an api for computers :P | 14:52 |
| lucasagomes | jroll, as well, but it does more than that | 14:52 |
| lucasagomes | for historical reasons | 14:52 |
| jroll | yeah | 14:52 |
| lucasagomes | if we had an abstraction above | 14:52 |
| lucasagomes | and nova would sit at the same level as ironic, but for vms | 14:52 |
| lucasagomes | that would be cool | 14:52 |
| *** mtanino has joined #openstack-ironic | 14:52 | |
| jroll | indeed | 14:52 |
| *** subscope has quit IRC | 14:53 | |
| devananda | dtantsur: metadata like driver_info -- the structure and contract is the API, not the data in it | 14:54 |
| dtantsur | devananda, I meant structure, yeah. adding a new driver -> new fields in driver_info -> version bump... | 14:55 |
| devananda | eg, how to determine what driver_info is required, whether it is valid, etc -- this is the API. but ^ no | 14:55 |
| dtantsur | adding new driver feature -> version bump | 14:55 |
| devananda | allowing a new field INSIDE driver_info -- not a bump | 14:55 |
| dtantsur | why? | 14:55 |
| devananda | as for vendor_passthru -- we (the Ironic project) have no control over what is or is not exposed in that | 14:56 |
| devananda | an out of tree driver can change the content of that endpoint when ever and how ever the want to | 14:56 |
| devananda | it is, essentially, unversionable | 14:56 |
| dtantsur | adding new HTTP endpoint results in bump, adding new field changing the behavior - no bump | 14:56 |
| devananda | or rather, its versions are not discoverable via version headers | 14:56 |
| dtantsur | devananda, in the ML you were against downstream changing API, not you said it's ok.... | 14:57 |
| devananda | that's how it is implemented, and in fact how it is designed | 14:57 |
| jroll | I really think anything that an out-of-tree driver can modify shouldn't require a bump. json fields, vendor passthru. | 14:57 |
| *** bethelwe_ has quit IRC | 14:57 | |
| devananda | dtantsur: correct. changing the REST API as defined by ironic (or any openstack project) is THE WORST THING EVER | 14:57 |
| *** bethelwell has joined #openstack-ironic | 14:58 | |
| dtantsur | so not out-of-tree vendor passthru? | 14:58 |
| dtantsur | * no | 14:58 |
| devananda | our API definition ends at the boundary of /vendor_passthru/ and /driver/vendor_passthru/ | 14:58 |
| dtantsur | why are you talking about them then? | 14:58 |
| devananda | what? | 14:58 |
| devananda | you brought them up earlier -- i'm responding | 14:58 |
| devananda | out of tree drivers can totally have vendor pasthru | 14:58 |
| dtantsur | so, in the ML I was assuming API is well API, and you assume that API means only some core endpoints, and nothing else | 14:59 |
| dtantsur | which is not something I agree with, but explains your (and IIRC Monty's) reaction to my words about downstream API | 14:59 |
| dtantsur | for me it's one more proof that this microversion suggestion is driver by us, the developers, not the users: the users are not aware of vendor-vs-non-vendor API thingy | 15:00 |
| dtantsur | at least users that I've seen | 15:00 |
| *** romainh has left #openstack-ironic | 15:01 | |
| jroll | to be fair, I'm not aware of any vendor passthru APIs in our tree, outside of deploy driver things | 15:01 |
| dtantsur | jroll, some DRAC stuff is proposed | 15:01 |
| dtantsur | jroll, and some iRMC as well | 15:02 |
| jroll | proposed, but not in our tree | 15:02 |
| jroll | so a user has never used vendor-passthru | 15:02 |
| jroll | because the only endpoints we ship are for internal use by ramdisks, not for users | 15:02 |
| dtantsur | which user? our do use DRAC passthru's... | 15:02 |
| jroll | well, that's not in tree, so I wasn't aware there was any installation with those available | 15:03 |
| jroll | do those get used? | 15:03 |
| dtantsur | yep | 15:03 |
| dtantsur | (that's why we added them, I'm not the biggest fan of downstream additions at all) | 15:03 |
| jroll | ok, great | 15:03 |
| jroll | but ironic, the open source project, has no control over whether that api breaks, has backward compat etc | 15:04 |
| dtantsur | exactly. but it's hard to explain to people, which we target this microversioning stuff at | 15:04 |
| *** ukalifon has quit IRC | 15:04 | |
| jroll | which is why we can only guarantee that vendor passthru is a thing, not that the endpoints there never break | 15:04 |
| jroll | is it? | 15:04 |
| jroll | "we are your vendor and added these things which are not versioned" | 15:05 |
| jroll | like, our user docs should have a better story for this and make it clear | 15:05 |
| dtantsur | maybe | 15:05 |
| dtantsur | but I still can't call it "we version the whole API" | 15:05 |
| dtantsur | I can call it "we version code API, where core API being <please tell me someone>" | 15:05 |
| jroll | well, you could version your vendor passthru if you wanted to, &version=3 :) | 15:06 |
| dtantsur | right, but I'm talking about microversion guidelines applied to Ironic | 15:06 |
| devananda | jroll: yes, there are in-tree vendor passthru | 15:06 |
| *** romainh has joined #openstack-ironic | 15:06 | |
| devananda | jroll: and have been for >1yr | 15:06 |
| devananda | look at the seamicro driver | 15:07 |
| jroll | right. we don't version the methods behind vendor passthru, only vendor passthru itself | 15:07 |
| *** romainh has left #openstack-ironic | 15:07 | |
| jroll | devananda: ok, good to know :) | 15:07 |
| *** subscope has joined #openstack-ironic | 15:07 | |
| devananda | they did vendor_passthru for network and SAn connectivity | 15:07 |
| devananda | loooong before ay other driver | 15:07 |
| *** trown|outttypeww is now known as trown | 15:07 | |
| jroll | neat | 15:08 |
| devananda | dtantsur: correct - we version the API contract, which includes the ROOT of "vendor_passthru" but not the content | 15:08 |
| devananda | that is not explicitly stated anywhere today, afaik, but should be | 15:08 |
| dtantsur | oh, it definitely should | 15:08 |
| devananda | ALSO - every service handles "but my driver is special!" in a different way | 15:08 |
| dtantsur | along with "we don't version JSON's, drivers, and only partly version provision states" | 15:08 |
| devananda | no other service has an explicit /vendor_passthru/ endpoint | 15:08 |
| devananda | a) I'm open to considering versioning some of the JSON | 15:09 |
| devananda | b) we should think about versioning the driver API | 15:09 |
| devananda | c) what about provision states is not versioned? | 15:09 |
| dtantsur | devananda, I'm thinking about one operator using new API and entering some new state, while the other still uses old API, but is fully affected by the new state | 15:10 |
| dtantsur | maybe I'm making up problems, not sure here | 15:11 |
| devananda | dtantsur: yea, I think that's a strawman | 15:11 |
| devananda | dtantsur: I don't disagree with the potential for it -- but I neither have any other way to do it, nor do I think it's really going to affect anyone | 15:11 |
| devananda | bu tmaybe I'm wron | 15:11 |
| dtantsur | devananda, fine, next question: what is justification for bumping microversion more often then every release? | 15:12 |
| dtantsur | (except for because we can, obviously :) | 15:12 |
| devananda | dtantsur: because people are very likely to deploy between releases | 15:13 |
| dtantsur | I know :) | 15:13 |
| devananda | folks want to believe (whether truthfully or not) that trunk can be deployed any time | 15:13 |
| *** dguerri is now known as dguerri` | 15:13 | |
| dtantsur | devananda, oh, I'm not talking about breaking changes | 15:13 |
| dtantsur | they can go out-of-band. what about the other. trunk is working, etc | 15:14 |
| *** jistr is now known as jistr|mtg | 15:17 | |
| *** Marga_ has joined #openstack-ironic | 15:17 | |
| *** Marga_ has quit IRC | 15:18 | |
| *** Marga_ has joined #openstack-ironic | 15:18 | |
| devananda | dtantsur: sorry, i dont understand teh quesiton | 15:19 |
| dtantsur | devananda, what's the point of bumping microversions for non-breaking changes between releases? | 15:20 |
| devananda | ah | 15:20 |
| dtantsur | tbh I'd prefer real-world use cases | 15:20 |
| devananda | here you go | 15:20 |
| devananda | imagine I write some tools that talk to ironic to do some things | 15:20 |
| devananda | and imagine there are multiple ironic clusters that I talk to -- but someone else manages those services | 15:21 |
| devananda | and a different persion manages each cluster (ie, region) | 15:21 |
| devananda | and there's an API change (non-breaking) rolled out in one of those regions | 15:21 |
| devananda | as a user, a) I can't detect this, and worse yet, b) if I start using it in region X, I have no way to detect whether it is supported in region Y | 15:22 |
| devananda | NOW | 15:22 |
| devananda | expressing this as a use-case makes the problem clear, at least fo rme | 15:22 |
| devananda | "as a user, I want to detect the capabilities of the API endpoint that I am talking to." And secondarily, "as a user, I want to determine whether it is different from any other API I am talking to" | 15:23 |
| dtantsur | cool, thanks! it makes a good case for feature-discovery API, but not for versions. versions are 1. more vague, 2. imply hard work for feature-gating things | 15:23 |
| devananda | version headers or URLs are only one way to do that -- but not hte only way | 15:23 |
| devananda | I totally think that our REST API needs to have a /schema/ endpoint | 15:23 |
| dtantsur | now I'll refine my question a bit: | 15:24 |
| devananda | that a user ca nquery to discover what the schema is, what resources or subresources are supported | 15:24 |
| devananda | and then compare that between clouds | 15:24 |
| devananda | now - that would be helfpul, but puts a lot of burden on clients, whereas just checking the version # is simpler, it's also less desccriptive | 15:24 |
| dtantsur | what's the point of gating feature behind API versions, i.e. disabling them based on user input? (provided that changes are backward compatible) | 15:24 |
| dtantsur | I was thinking about $ ironic feature-supported inspection -> true/false | 15:25 |
| dtantsur | or kind of this | 15:25 |
| dtantsur | which is explicit and does not make us go into troubles with disabling features depending on version | 15:25 |
| devananda | consistency | 15:26 |
| devananda | as a user, I have written my tool expecting a particular API | 15:26 |
| dtantsur | hmm? what's the use case of such consistency? | 15:26 |
| devananda | I want my cloud to behave the same way -- or tell me that it can not | 15:27 |
| dtantsur | it will behave the same - otherwise it's a breaking change | 15:27 |
| dtantsur | how will addition of inspection break you flows unaware of it? | 15:27 |
| devananda | dtantsur: see my use cases above | 15:30 |
| devananda | dtantsur: adding a new API doesn't break current users, if you think in terms of only one cloud | 15:30 |
| dtantsur | devananda, you have talked about feature discovery, but not about feature gating. I don't see these 2 as one problem | 15:30 |
| devananda | dtantsur: but it has very real chances to break someone if they are using multiple clouds | 15:30 |
| dtantsur | how? any real world example? | 15:31 |
| devananda | I already explained it | 15:31 |
| devananda | shall I paste? | 15:31 |
| *** Marga_ has quit IRC | 15:31 | |
| *** coolsvap is now known as coolsvap|afk | 15:31 | |
| dtantsur | devananda, imagine we add HTTP API `/v1/nodes/UUID/kittens`, imaging it's mad discoverable by the feature-discovery API | 15:31 |
| devananda | what's the difference between "feature discover" and "feature gating" in your mind? | 15:31 |
| dtantsur | discover = I'm aware that feature is present. gating = feature is disabled unless I send magic header | 15:32 |
| dtantsur | difference is in burden that is put on developers and reviewers every time to make sure gating is present and correct | 15:32 |
| devananda | dtantsur: what is difference to a user? | 15:34 |
| dtantsur | devananda, my opinion is that once feature discovery is provided, user does not care about feature gating. so not difference for a user. | 15:34 |
| dtantsur | (that's still about compatible changes) | 15:35 |
| *** jistr|mtg is now known as jistr | 15:35 | |
| devananda | dtantsur: so, as a user of >1 cloud, I want to build an app on cloud X, and know that it will work on cloud Y | 15:35 |
| devananda | dtantsur: feature discovery does not provide ^ | 15:35 |
| devananda | unless I, as a user, also write some method to cache and compare schemas/feature-lists/etc between clouds | 15:36 |
| dtantsur | devananda, it's provided by backward compatibility guarantee | 15:36 |
| devananda | no | 15:36 |
| devananda | it's not | 15:36 |
| dtantsur | then it's not compatibility | 15:36 |
| devananda | if cloud X has your kittens extension and cloud Y doe snot | 15:36 |
| devananda | then my app will fail on cloud Y | 15:36 |
| devananda | even you though you added kittens in a "backwards compatible way" | 15:36 |
| dtantsur | just like with feature gating... | 15:36 |
| dtantsur | if you don't use feature discovery, your app will still fail, but with a different error | 15:37 |
| dtantsur | "API version x.y is not supported" | 15:37 |
| devananda | correct | 15:37 |
| dtantsur | BUT! | 15:37 |
| devananda | which is a heck of a lot better than blowing up half way through some workflow | 15:37 |
| devananda | when it fails to access /kittehs/ | 15:37 |
| dtantsur | if you use feature discovery, you will know even before you try | 15:37 |
| devananda | and if you bump the api version when you add /kittehs/, same thing -- it works | 15:38 |
| devananda | makes me 9as the user) know that something changed | 15:38 |
| *** zz_natorious is now known as natorious | 15:38 | |
| dtantsur | again, use feature discovery, no problem | 15:39 |
| dtantsur | you will give even better message to a user: cloud X does not support kittens | 15:39 |
| *** hemnafk is now known as hemna | 15:40 | |
| dtantsur | of good things, it will cover stuff like cleaning, which is not covered by API directly | 15:40 |
| dtantsur | e.g. you'll be able to figure out if cleaning is gonna work on your cloud | 15:40 |
| dtantsur | how to do it with microversions, provided that cleaning is (or is not) triggered automatically? | 15:40 |
| *** natorious is now known as zz_natorious | 15:42 | |
| dprince | Heads up on this Nova issue which breaks the ephemeral partition w/ Ironic: https://bugs.launchpad.net/tripleo/+bug/1464239 | 15:42 |
| openstack | Launchpad bug 1464239 in tripleo "mount: special device /dev/sdb does not exist" [Critical,In progress] - Assigned to Dan Prince (dan-prince) | 15:42 |
| *** coolsvap|afk is now known as coolsvap | 15:42 | |
| openstackgerrit | Devananda van der Veen proposed openstack/ironic: remove unneeded sqlalchemy-migrate requirement https://review.openstack.org/190696 | 15:45 |
| devananda | cinerama: ^ | 15:45 |
| devananda | TheJulia: ^ | 15:46 |
| TheJulia | ack | 15:46 |
| devananda | dropping offline now. out of battery and its time for drinks/dinner | 15:46 |
| TheJulia | 10-4 | 15:46 |
| devananda | dtantsur: so we dont have feature discovery today. It would be helpful if we did | 15:47 |
| * lucasagomes reads the whole scrollback... | 15:47 | |
| * devananda is now afk | 15:47 | |
| cinerama | devananda: thanks for clearing that up | 15:48 |
| dtantsur | devananda, and my suggestion is to do it (which is very easy) instead of feature gate based on versions (which is hard) | 15:49 |
| dtantsur | devananda, hope I did bring some useful feedback :) | 15:49 |
| *** saripurigopi has joined #openstack-ironic | 15:52 | |
| *** ijw has joined #openstack-ironic | 15:56 | |
| *** lazy_prince has quit IRC | 15:57 | |
| *** coolsvap is now known as coolsvap|afk | 15:58 | |
| *** achanda has joined #openstack-ironic | 15:58 | |
| openstackgerrit | John Trowbridge proposed stackforge/ironic-discoverd: Rename edeploy plugin https://review.openstack.org/189944 | 16:08 |
| openstackgerrit | John Trowbridge proposed stackforge/ironic-discoverd: [WIP] Fix edeploy plugin puts too much data in Ironic extra column https://review.openstack.org/189427 | 16:08 |
| *** Marga_ has joined #openstack-ironic | 16:09 | |
| dtantsur | g'night | 16:11 |
| *** dtantsur is now known as dtantsur|afk | 16:11 | |
| NobodyCam | nighth dtantsur|afk | 16:11 |
| lucasagomes | dtantsur|afk, night | 16:15 |
| lucasagomes | dprince, ew, thanks for the heads up | 16:16 |
| * lucasagomes looks at the change in nova, seems ok-ish | 16:17 | |
| *** achanda has quit IRC | 16:17 | |
| lucasagomes | folks, since microversion was the talk. If you have a time mind taking a look at https://review.openstack.org/#/c/183491/ | 16:18 |
| lucasagomes | (that bumps the micro version) | 16:18 |
| *** zz_natorious is now known as natorious | 16:19 | |
| *** mgoddard has quit IRC | 16:19 | |
| *** mgoddard has joined #openstack-ironic | 16:20 | |
| *** krtaylor has quit IRC | 16:22 | |
| jlvillal | lucasagomes: Is anyone working on the microversion error body? As discussed on the mailing list. | 16:24 |
| jlvillal | About sending down 'versionFault' | 16:24 |
| lucasagomes | jlvillal, not that I am aware of | 16:25 |
| lucasagomes | jlvillal, I would wait to the API group guidelines to do it | 16:25 |
| jlvillal | lucasagomes: Okay. I'll keep my eye on: https://review.openstack.org/#/c/187112/ | 16:25 |
| lucasagomes | jlvillal, cool, thanks! | 16:26 |
| *** e0ne has joined #openstack-ironic | 16:27 | |
| *** r-daneel has joined #openstack-ironic | 16:27 | |
| *** subscope has quit IRC | 16:27 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: When boot option is not persisted, set boot on next power on https://review.openstack.org/177642 | 16:27 |
| *** achanda has joined #openstack-ironic | 16:35 | |
| *** jcoufal has quit IRC | 16:36 | |
| *** Nisha has joined #openstack-ironic | 16:40 | |
| *** saripurigopi has quit IRC | 16:46 | |
| *** Marga_ has quit IRC | 16:47 | |
| *** saripurigopi has joined #openstack-ironic | 16:49 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Wake-On-Lan Power interface https://review.openstack.org/179078 | 16:50 |
| *** ndipanov has quit IRC | 16:50 | |
| lucasagomes | folks I will call it a day | 16:51 |
| *** Marga_ has joined #openstack-ironic | 16:51 | |
| lucasagomes | have a good night everyone | 16:51 |
| *** sambetts has quit IRC | 16:51 | |
| *** derekh has quit IRC | 16:51 | |
| NobodyCam | night lucasagomes | 16:51 |
| lucasagomes | g'night | 16:51 |
| *** sambetts has joined #openstack-ironic | 16:54 | |
| *** dprince has quit IRC | 16:56 | |
| *** achanda has quit IRC | 16:56 | |
| *** bethelwell has quit IRC | 16:56 | |
| *** saripurigopi has quit IRC | 17:00 | |
| *** lazy_prince has joined #openstack-ironic | 17:02 | |
| *** bnemec has joined #openstack-ironic | 17:06 | |
| *** lucasagomes has quit IRC | 17:07 | |
| *** e0ne has quit IRC | 17:09 | |
| *** spandhe has joined #openstack-ironic | 17:13 | |
| *** lazy_prince has quit IRC | 17:14 | |
| openstackgerrit | Josh Gachnang proposed openstack/ironic: Add additional logging around cleaning https://review.openstack.org/190728 | 17:20 |
| *** gridinv_ has joined #openstack-ironic | 17:20 | |
| *** gridinv_ has quit IRC | 17:33 | |
| *** jistr has quit IRC | 17:35 | |
| *** athomas has quit IRC | 17:36 | |
| openstackgerrit | Jarrod Johnson proposed stackforge/pyghmi: Handle Lenovo Thinkserver lacking OEM inventory https://review.openstack.org/190733 | 17:38 |
| *** dprince has joined #openstack-ironic | 17:41 | |
| *** pcrews has quit IRC | 17:42 | |
| *** radek_ has joined #openstack-ironic | 17:44 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic-specs: Update network provider spec https://review.openstack.org/187829 | 17:44 |
| jroll | wheeeeeeeeee | 17:44 |
| jlvillal | BadCub: Did you get a consensus for the next sprint date? | 17:50 |
| *** ifarkas has quit IRC | 17:51 | |
| openstackgerrit | Merged stackforge/pyghmi: Handle Lenovo Thinkserver lacking OEM inventory https://review.openstack.org/190733 | 17:56 |
| *** achanda has joined #openstack-ironic | 17:57 | |
| *** Marga_ has quit IRC | 17:59 | |
| *** Marga_ has joined #openstack-ironic | 17:59 | |
| *** hemna is now known as hemnafk | 18:01 | |
| *** degorenko has quit IRC | 18:01 | |
| BadCub | jlvillal: so far the responses indicate the proposed date Aug 12-14 | 18:02 |
| jlvillal | BadCub: Thanks. That date I can make :) | 18:02 |
| BadCub | 10 Yes for that date | 18:03 |
| *** Nisha has quit IRC | 18:05 | |
| *** pelix has quit IRC | 18:11 | |
| *** krtaylor has joined #openstack-ironic | 18:18 | |
| *** Marga_ has quit IRC | 18:19 | |
| *** Marga_ has joined #openstack-ironic | 18:19 | |
| *** pcrews has joined #openstack-ironic | 18:20 | |
| *** spandhe has quit IRC | 18:26 | |
| *** spandhe has joined #openstack-ironic | 18:29 | |
| *** achanda has quit IRC | 18:31 | |
| *** Marga_ has quit IRC | 18:32 | |
| *** Marga_ has joined #openstack-ironic | 18:32 | |
| openstackgerrit | Jarrod Johnson proposed stackforge/pyghmi: Gather Thinkserver CPU and Memory independently https://review.openstack.org/190749 | 18:34 |
| *** dprince has quit IRC | 18:34 | |
| *** achanda has joined #openstack-ironic | 18:37 | |
| *** hemnafk is now known as hemna | 18:43 | |
| openstackgerrit | John Trowbridge proposed stackforge/ironic-discoverd: Rename edeploy plugin https://review.openstack.org/189944 | 18:45 |
| openstackgerrit | John Trowbridge proposed stackforge/ironic-discoverd: Fix edeploy plugin puts too much data in Ironic extra column https://review.openstack.org/189427 | 18:45 |
| *** ijw has quit IRC | 18:45 | |
| *** trown is now known as trown|brb | 18:49 | |
| *** achanda has quit IRC | 18:50 | |
| *** e0ne has joined #openstack-ironic | 18:51 | |
| *** e0ne is now known as e0ne_ | 18:51 | |
| *** achanda has joined #openstack-ironic | 18:52 | |
| openstackgerrit | Josh Gachnang proposed openstack/ironic: Add additional logging around cleaning https://review.openstack.org/190728 | 18:53 |
| *** e0ne_ is now known as e0ne | 18:53 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Minor typo correction in new inventory module https://review.openstack.org/190760 | 18:55 |
| *** e0ne is now known as e0ne_ | 19:00 | |
| *** achanda has quit IRC | 19:00 | |
| *** achanda has joined #openstack-ironic | 19:02 | |
| *** bkero has quit IRC | 19:04 | |
| *** ijw has joined #openstack-ironic | 19:05 | |
| *** greghaynes has quit IRC | 19:05 | |
| *** greghaynes has joined #openstack-ironic | 19:05 | |
| *** bkero has joined #openstack-ironic | 19:05 | |
| openstackgerrit | Merged openstack/bifrost: Initial inventory module https://review.openstack.org/184084 | 19:16 |
| *** trown|brb is now known as trown | 19:17 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Change pip install method and add six to requirements https://review.openstack.org/190768 | 19:20 |
| *** e0ne_ is now known as e0ne | 19:35 | |
| NobodyCam | brb | 19:35 |
| *** Marga_ has quit IRC | 19:36 | |
| *** Marga_ has joined #openstack-ironic | 19:37 | |
| *** achanda has quit IRC | 19:41 | |
| *** hemna is now known as hemnafood | 19:44 | |
| *** dguerri` is now known as dguerri | 19:44 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Correct CSV processing in inventory https://review.openstack.org/190781 | 19:45 |
| TheJulia | NobodyCam: ^^^ | 19:45 |
| *** pcrews has quit IRC | 19:47 | |
| *** dguerri is now known as dguerri` | 19:49 | |
| openstackgerrit | Merged openstack/bifrost: Minor typo correction in new inventory module https://review.openstack.org/190760 | 20:05 |
| *** thrash is now known as thrash|g0ne | 20:11 | |
| rloo | hey NobodyCam or BadCub or even better if devananda is around: this BP has been approved: https://blueprints.launchpad.net/ironic/+spec/new-boot-interface | 20:15 |
| rloo | is devananda the only one that can approve BPs? | 20:15 |
| NobodyCam | let me try | 20:15 |
| NobodyCam | rloo: hows that look? | 20:16 |
| NobodyCam | seemed to work for me | 20:16 |
| rloo | NobodyCam: awesome. thx. would you also be so kind as to set the URL for the specification? | 20:17 |
| BadCub | rloo: NobodyCam this BP is targeted for L I assume? | 20:18 |
| NobodyCam | I think* I set that | 20:18 |
| rloo | BadCub: yuppers | 20:18 |
| NobodyCam | please dbl check me on that | 20:18 |
| BadCub | I set the milestone on it for L1 | 20:19 |
| NobodyCam | oh ya I didn;t set that | 20:19 |
| BadCub | set the priority as medium as well | 20:19 |
| *** gridinv_ has joined #openstack-ironic | 20:19 | |
| rloo | thx BadCub (i'm fine with that cuz i don't really know and it doesn't really matter now i don't think) | 20:20 |
| BadCub | rloo: coolness :) | 20:21 |
| BadCub | IT can always be changed as well, if needed | 20:21 |
| NobodyCam | rloo: set url :) | 20:21 |
| rloo | thx NobodyCam and BadCub | 20:22 |
| NobodyCam | :) | 20:22 |
| BadCub | rloo: Yw :) | 20:22 |
| rloo | btw NobodyCam, why do we point to github and not the specs url? | 20:22 |
| rloo | NobodyCam: i mean http://specs.openstack.org/openstack/ironic-specs/specs/liberty/new-boot-interface.html | 20:23 |
| NobodyCam | ?? i can set to that... I tend to read in github so just default for me | 20:23 |
| BadCub | NobodyCam: we usually use ^ on the set URL | 20:24 |
| rloo | NobodyCam: ahh. i think i've seen urls to github or openstack's git (which is probably preferred over github). | 20:24 |
| rloo | NobodyCam: i tend to look at the specs from specs.openstack.org. Anything is better than nothing though. | 20:25 |
| NobodyCam | rloo: updated :) | 20:25 |
| *** Sukhdev has joined #openstack-ironic | 20:25 | |
| rloo | thx NobodyCam! | 20:26 |
| *** spandhe has quit IRC | 20:27 | |
| NobodyCam | *blush* :) | 20:27 |
| *** spandhe has joined #openstack-ironic | 20:28 | |
| *** spandhe has quit IRC | 20:28 | |
| openstackgerrit | Merged openstack/bifrost: Initial dynamic enrollment role https://review.openstack.org/188176 | 20:29 |
| openstackgerrit | Merged openstack/bifrost: Enable agent_ipmitool driver when testing https://review.openstack.org/188623 | 20:29 |
| *** pcrews has joined #openstack-ironic | 20:30 | |
| *** achanda has joined #openstack-ironic | 20:31 | |
| *** alexpilotti has quit IRC | 20:31 | |
| *** gridinv_ has quit IRC | 20:32 | |
| *** gridinv_ has joined #openstack-ironic | 20:32 | |
| *** ijw has quit IRC | 20:35 | |
| *** ijw has joined #openstack-ironic | 20:36 | |
| *** dguerri` is now known as dguerri | 20:39 | |
| openstackgerrit | Merged stackforge/pyghmi: Gather Thinkserver CPU and Memory independently https://review.openstack.org/190749 | 20:41 |
| *** achanda has quit IRC | 20:43 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Add initial dynamic deployment role https://review.openstack.org/188773 | 20:44 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Change pip install method and add six to requirements https://review.openstack.org/190768 | 20:47 |
| openstackgerrit | Merged openstack/bifrost: Addition of dynamic configuration drives https://review.openstack.org/188624 | 20:50 |
| *** e0ne has quit IRC | 20:54 | |
| *** Marga_ has quit IRC | 20:58 | |
| *** Marga_ has joined #openstack-ironic | 20:59 | |
| -openstackstatus- NOTICE: Gerrit has been restarted to terminate a persistent looping third-party CI bot | 21:07 | |
| *** trown is now known as trown|outttypeww | 21:13 | |
| *** ijw has quit IRC | 21:19 | |
| *** ijw has joined #openstack-ironic | 21:19 | |
| *** hemnafood is now known as hemna | 21:21 | |
| *** dguerri is now known as dguerri` | 21:21 | |
| *** radek_ has quit IRC | 21:24 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic-specs: Update network provider spec https://review.openstack.org/187829 | 21:26 |
| openstackgerrit | Merged openstack/bifrost: Add initial dynamic deployment role https://review.openstack.org/188773 | 21:28 |
| *** spandhe has joined #openstack-ironic | 21:30 | |
| openstackgerrit | Merged openstack/bifrost: Correct mistakes in license notices https://review.openstack.org/189079 | 21:33 |
| *** spandhe_ has joined #openstack-ironic | 21:35 | |
| *** ijw_ has joined #openstack-ironic | 21:36 | |
| *** spandhe has quit IRC | 21:37 | |
| *** spandhe_ is now known as spandhe | 21:37 | |
| mrda | Morning Ironic | 21:37 |
| *** ijw has quit IRC | 21:39 | |
| rloo | morning mrda. I was *just* thinking that I ought to look at your patch :) | 21:42 |
| mrda | hi rloo - thanks :) | 21:43 |
| NobodyCam | mornign mrda | 21:43 |
| mrda | hi NobodyCam | 21:43 |
| *** natorious is now known as zz_natorious | 21:46 | |
| rloo | hey NobodyCam, this BP should be approved too, the spec was approved: https://blueprints.launchpad.net/ironic/+spec/version-caching | 21:47 |
| mrda | rloo: I really do appreciate your follow through :) | 21:50 |
| *** ijw_ has quit IRC | 21:52 | |
| *** jjohnson2 has quit IRC | 21:54 | |
| openstackgerrit | Christopher Dearborn proposed openstack/ironic: Fix failing unit tests under py34 https://review.openstack.org/190835 | 21:55 |
| rloo | mrda: honestly, i was procrastinating. now i'm not sure i'll be able to get it done today. will see... | 21:57 |
| *** [1]cdearborn has joined #openstack-ironic | 21:59 | |
| mrda | rloo: there's no actualy change to the code since you last reviewed it. It was only the commit message that changed the blueprint reference as per your last review comment | 22:00 |
| rloo | mrda: ahhh, but you're assuming I had reviewed up to there. I hadn't :-( | 22:01 |
| mrda | (but if you want to review it all, it would be welcoemd :) | 22:01 |
| mrda | Do it tomorrow then :) | 22:02 |
| rloo | mrda: not here tomorrow. either tonight or monday | 22:02 |
| mrda | It's nice to get a break - it can wait, I don't think anyone is screaming for this change | 22:03 |
| *** ijw has joined #openstack-ironic | 22:04 | |
| *** mitchjameson has joined #openstack-ironic | 22:04 | |
| rloo | mrda: yeah, nothing (in ironic) so far is urgent. just that it was on my list to do this week, will see... | 22:07 |
| *** cdearborn has quit IRC | 22:07 | |
| *** [1]cdearborn has quit IRC | 22:18 | |
| openstackgerrit | Merged openstack/ironic-python-agent: Updated from global requirements https://review.openstack.org/190427 | 22:22 |
| *** absubram has quit IRC | 22:22 | |
| *** Sukhdev has quit IRC | 22:25 | |
| NobodyCam | rloo: updated :) | 22:27 |
| *** Nisha has joined #openstack-ironic | 22:30 | |
| mrda | thanks NobodyCam | 22:33 |
| *** bnemec has quit IRC | 22:34 | |
| NobodyCam | :) | 22:36 |
| *** Nisha has quit IRC | 22:36 | |
| *** jgrimm is now known as zz_jgrimm | 22:51 | |
| *** krtaylor has quit IRC | 23:04 | |
| *** ijw has quit IRC | 23:05 | |
| *** zhenguo has joined #openstack-ironic | 23:15 | |
| *** Sukhdev has joined #openstack-ironic | 23:19 | |
| *** yuanying has joined #openstack-ironic | 23:23 | |
| *** chlong has joined #openstack-ironic | 23:35 | |
| * NobodyCam calls it a early day | 23:38 | |
| *** hemna is now known as hemnafk | 23:45 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!