| *** openstack has joined #openstack-ironic | 15:45 | |
| *** mjturek1 has joined #openstack-ironic | 15:46 | |
| *** Sukhdev_ has joined #openstack-ironic | 15:46 | |
| openstackgerrit | Merged openstack/ironic: force releasenotes warnings to be treated as errors https://review.openstack.org/250572 | 15:46 |
|---|---|---|
| *** sturivnyi has joined #openstack-ironic | 15:48 | |
| *** e0ne has joined #openstack-ironic | 15:48 | |
| sambetts | p/ mariojv | 15:49 |
| sambetts | o/ | 15:49 |
| *** dims has joined #openstack-ironic | 15:49 | |
| *** electrofelix has joined #openstack-ironic | 15:51 | |
| *** smoriya_afk has joined #openstack-ironic | 15:51 | |
| *** smoriya_afk is now known as smoriya | 15:51 | |
| *** amotoki has quit IRC | 15:53 | |
| *** jistr has quit IRC | 15:57 | |
| NobodyCam | good morning Ironicers | 15:57 |
| BadCub | morning folks | 15:57 |
| *** lucas-afk is now known as lucasagomes | 15:58 | |
| lucasagomes | TheJulia, NobodyCam morning | 15:58 |
| lucasagomes | mariojv, morning too :-) | 15:58 |
| NobodyCam | hey hey lucasagomes :) | 15:58 |
| dtantsur | morning NobodyCam, BadCub | 15:58 |
| NobodyCam | morning dtantsur | 15:58 |
| *** jasondotstar has joined #openstack-ironic | 15:58 | |
| openstackgerrit | Imre Farkas proposed openstack/ironic: DRAC: switch to python-dracclient on management interface https://review.openstack.org/249256 | 15:58 |
| NobodyCam | morning to rloo jroll devananda sambetts jlvillal and all oters not listed here | 15:59 |
| jlvillal | NobodyCam, morning! | 16:00 |
| mgould | NobodyCam, morning! | 16:00 |
| *** e0ne has quit IRC | 16:02 | |
| rloo | ohhh, yawn, stretch, good morning NobodyCam, BadCub, lucasagomes, TheJulia, dtantsur, mariojv, jlvillal, mgould, ifarkas, jroll :) | 16:02 |
| lucasagomes | rloo, morning! | 16:02 |
| vdrok | morning mgould jlvillal mariojv NobodyCam BadCub TheJulia rloo ! | 16:02 |
| dtantsur | morning rloo :) | 16:03 |
| *** nicodemos has joined #openstack-ironic | 16:03 | |
| jlvillal | vdrok, :) | 16:04 |
| ifarkas | morning rloo! | 16:04 |
| vdrok | morning ifarkas :) | 16:04 |
| rloo | hi vdrok! | 16:04 |
| ifarkas | morning all :-) | 16:05 |
| *** kevinbenton_ has quit IRC | 16:07 | |
| *** mitchjameson has joined #openstack-ironic | 16:07 | |
| NobodyCam | :) | 16:07 |
| *** kevinbenton has joined #openstack-ironic | 16:07 | |
| *** Marga_ has joined #openstack-ironic | 16:11 | |
| *** Marga_ has quit IRC | 16:13 | |
| *** Marga_ has joined #openstack-ironic | 16:14 | |
| *** Nisha has joined #openstack-ironic | 16:15 | |
| *** arif-ali_ has joined #openstack-ironic | 16:15 | |
| *** arif-ali has quit IRC | 16:15 | |
| *** openstackstatus has joined #openstack-ironic | 16:15 | |
| *** ChanServ sets mode: +v openstackstatus | 16:15 | |
| *** arif-ali_ is now known as arif-ali | 16:15 | |
| sambetts | o/ NobodyCam rloo ifarkas | 16:15 |
| NobodyCam | :) | 16:20 |
| rloo | dtantsur: wrt https://review.openstack.org/#/c/250745/, does it need a reno? | 16:22 |
| dtantsur | probably? I'm still not sure myself :) | 16:23 |
| dtantsur | jroll, wdyt ^^ | 16:23 |
| rloo | dtantsur: it is a 'high' and critical for tripleo so I am guessing yes. | 16:23 |
| dtantsur | yeah, I'll update it | 16:23 |
| rloo | dtantsur: ok thx. ping me and i'll review it then :) | 16:23 |
| openstackgerrit | Sergii Turivnyi proposed openstack/python-ironicclient: Tests for testing node-delete command https://review.openstack.org/251458 | 16:26 |
| *** pas-ha has quit IRC | 16:27 | |
| *** pas-ha has joined #openstack-ironic | 16:27 | |
| openstackgerrit | Sergii Turivnyi proposed openstack/python-ironicclient: Add tests for testing node-delete command https://review.openstack.org/251458 | 16:28 |
| devananda | g'morning, all | 16:29 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Fall back to old boot.ipxe behaviour if inc command is not found https://review.openstack.org/250745 | 16:29 |
| dtantsur | rloo, lucasagomes ^^ | 16:29 |
| dtantsur | morning devananda | 16:29 |
| NobodyCam | morning devananda | 16:29 |
| lucasagomes | devananda, morning | 16:29 |
| lucasagomes | dtantsur, will take a look | 16:29 |
| rloo | thx dtantsur, looking | 16:29 |
| rloo | morning devananda! | 16:29 |
| *** baoli has quit IRC | 16:29 | |
| *** baoli has joined #openstack-ironic | 16:31 | |
| *** aarefiev22 has quit IRC | 16:32 | |
| TheJulia | good morning devananda | 16:32 |
| vdrok | devananda, morning | 16:33 |
| mgould | devananda, morning! | 16:33 |
| *** chlong has joined #openstack-ironic | 16:40 | |
| *** achanda_ has quit IRC | 16:41 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Fall back to old boot.ipxe behaviour if inc command is not found https://review.openstack.org/250745 | 16:44 |
| *** mgoddard_ has joined #openstack-ironic | 16:44 | |
| dtantsur | rloo, fixed ^^ thanks | 16:45 |
| rloo | dtantsur: ah, ok, looking | 16:45 |
| rloo | ding, ding, 15 minutes to our meeting | 16:46 |
| * TheJulia goes and makes coffee | 16:46 | |
| openstackgerrit | Imre Farkas proposed openstack/ironic: DO NOT MERGE - disable ssl warnings in tests https://review.openstack.org/251465 | 16:47 |
| *** baoli has quit IRC | 16:48 | |
| *** mgoddard__ has quit IRC | 16:48 | |
| *** baoli has joined #openstack-ironic | 16:50 | |
| *** purp has quit IRC | 16:51 | |
| *** cinerama has joined #openstack-ironic | 16:51 | |
| *** sacharya has joined #openstack-ironic | 16:52 | |
| *** rpioso has joined #openstack-ironic | 16:52 | |
| *** purp has joined #openstack-ironic | 16:54 | |
| *** ChubYann has joined #openstack-ironic | 16:55 | |
| *** sacharya has quit IRC | 16:56 | |
| *** sacharya has joined #openstack-ironic | 16:58 | |
| *** achanda has joined #openstack-ironic | 16:59 | |
| *** athomas has joined #openstack-ironic | 16:59 | |
| *** ukalifon has quit IRC | 16:59 | |
| * devananda heads over to the meeting room | 17:00 | |
| *** cdearborn has joined #openstack-ironic | 17:01 | |
| *** coolsvap has quit IRC | 17:01 | |
| *** mgoddard_ has quit IRC | 17:02 | |
| *** mitchjameson has quit IRC | 17:02 | |
| *** mgoddard has joined #openstack-ironic | 17:02 | |
| openstackgerrit | Brad P. Crochet proposed openstack/ironic-specs: OpenStackClient plugin spec https://review.openstack.org/247539 | 17:03 |
| jroll | rloo: dtantsur: if we have to ask, it probably needs a note :) | 17:04 |
| dtantsur | jroll, already done, thanks :) | 17:04 |
| rloo | jroll: yeah :) | 17:04 |
| *** coolsvap has joined #openstack-ironic | 17:05 | |
| *** derekh has joined #openstack-ironic | 17:05 | |
| *** Sukhdev_ has quit IRC | 17:06 | |
| *** krtaylor has joined #openstack-ironic | 17:06 | |
| *** yuikotakada has joined #openstack-ironic | 17:07 | |
| *** yuikotakada has quit IRC | 17:09 | |
| *** yuikotak_ has joined #openstack-ironic | 17:10 | |
| *** Sukhdev_ has joined #openstack-ironic | 17:12 | |
| *** Sukhdev_ has quit IRC | 17:12 | |
| *** boris-42_ has quit IRC | 17:15 | |
| *** rebase has joined #openstack-ironic | 17:19 | |
| *** jobewan has joined #openstack-ironic | 17:22 | |
| *** Piet has joined #openstack-ironic | 17:22 | |
| dtantsur | that's how we fail to encourage people to use the ML. it's really weird. | 17:22 |
| *** thingee has joined #openstack-ironic | 17:29 | |
| *** sacharya has quit IRC | 17:31 | |
| openstackgerrit | Merged openstack/ironic-python-agent: Updated from global requirements https://review.openstack.org/250955 | 17:35 |
| thingee | jroll: just a reminder that communication for third party ci's are going out today. | 17:38 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-python-agent: Support Linux-IO in addition to tgtd https://review.openstack.org/234861 | 17:38 |
| thingee | jroll: did we by chance appoint people who can answer questions in channel in between third party ci help meetings? | 17:38 |
| rloo | thingee: jroll is on vacation this week. he was here a few minutes ago, not sure when he'll be back. | 17:39 |
| rloo | krtaylor: ^^ do you know the answers to thingee's questions? | 17:40 |
| *** ifarkas has quit IRC | 17:41 | |
| krtaylor | thingee, I can answer questions | 17:42 |
| *** paul-carlton1 has joined #openstack-ironic | 17:43 | |
| thingee | rloo krtaylor thanks! I'll send an email to jroll as a last chance to stop things. The plan we had from last wednesday was to communicate to vendors the information from the third party ci spec. | 17:44 |
| *** Nisha has quit IRC | 17:44 | |
| thingee | and that would go out today | 17:44 |
| krtaylor | thingee, thanks for the communication email, huge help, can you send me the template for future reference? | 17:44 |
| rloo | thingee: it should go out today. I don't think there's any reason to not do it. | 17:44 |
| krtaylor | ++ | 17:44 |
| *** Marga_ has quit IRC | 17:45 | |
| thingee | krtaylor: https://etherpad.openstack.org/p/IronicCI | 17:47 |
| thingee | krtaylor: it's at the bottom | 17:47 |
| krtaylor | thingee, thanks | 17:48 |
| *** ionutbalutoiu has joined #openstack-ironic | 17:49 | |
| anteaya | thingee: can you clarify the use of the third party systems wikipage please | 17:51 |
| *** penick has joined #openstack-ironic | 17:51 | |
| anteaya | thingee: there seems to be some confusion | 17:52 |
| thingee | anteaya: sure, what needs to be fixed? | 17:52 |
| *** hblixt has joined #openstack-ironic | 17:53 | |
| anteaya | thingee: third party ci systems are to use the wikipage, not to announce themselves on the mailing list | 17:53 |
| anteaya | thingee: I would like to ensure that information is clear | 17:53 |
| dtantsur | the first time I see a nearly "shut up" response on the ML (twice for a thread) | 17:54 |
| *** achanda has quit IRC | 17:54 | |
| dtantsur | could you please not spam our channel with your discussions? | 17:54 |
| anteaya | sure | 17:56 |
| anteaya | I think you are not understanding the extent to which the openstack community has already had to deal with third party things | 17:57 |
| dtantsur | nobody cared to explain | 17:57 |
| anteaya | and expectations that are in place around that | 17:57 |
| anteaya | I am explaining | 17:58 |
| *** [1]cdearborn has joined #openstack-ironic | 17:58 | |
| anteaya | as a development communitity we have had to deal with an onslaught of activity from myriad third party ci systems for over 2 years | 17:58 |
| anteaya | they don't pay attention to communitiy events other than their driver | 17:59 |
| anteaya | and they dont' pay attention to weekly project meetings | 17:59 |
| anteaya | there are some exceptions to this | 17:59 |
| anteaya | and that is wonderful | 17:59 |
| anteaya | but the majority pay attention to only their driver | 17:59 |
| *** paul-carlton1 has quit IRC | 18:00 | |
| anteaya | and spam many lines of communication and the developers don't like it | 18:00 |
| devananda | derekh: happy to continue here | 18:00 |
| anteaya | which is why certain things were set up, to deal with the spamming and to minimize the disruption to devs | 18:00 |
| anteaya | one of these was the third party systems wikipage | 18:00 |
| derekh | devananda: I gotta run, but can continue tomorrow or on the mail thread if you want | 18:01 |
| *** yuikotak_ has quit IRC | 18:01 | |
| anteaya | which was specifically set up so that anyone can check the existance of a third party ci system | 18:01 |
| dtantsur | anteaya, while I (now) understand and share your motives, I still don't think the response was fair. It was we, the ironic cores, who forced driver folks to set up the CI. a simple bit of explanation *why* they should not announce that they did what we asked them to would be very beneficial for future collaborations with them | 18:02 |
| anteaya | that is fine | 18:02 |
| *** derekh has quit IRC | 18:02 | |
| anteaya | many things I have done and continue to do in the third party space don't seem fair until the big picture is understood | 18:02 |
| *** penick_ has joined #openstack-ironic | 18:02 | |
| anteaya | dtantsur: that is fine, I asked who told them to announce on the mailing list and was told that noone from ironic has done so | 18:03 |
| devananda | penick: ohhai. have some time today to discuss the claims api work? | 18:03 |
| anteaya | dtantsur: have you told anyone to announce on the mailing list? | 18:04 |
| *** penick has quit IRC | 18:04 | |
| *** penick_ is now known as penick | 18:04 | |
| dtantsur | anteaya, if I was them, I would treat it as obvious. if they asked me, I would say "definitely on the ML" | 18:04 |
| anteaya | please don't do that in future | 18:04 |
| dtantsur | they didn't, but I would definitely tell them to do before this conversation | 18:04 |
| anteaya | have you done so in the past? | 18:04 |
| dtantsur | I *think* never | 18:04 |
| anteaya | great, so hopefully you won't now | 18:04 |
| anteaya | wonderful | 18:04 |
| * rloo would have also thought to mention to someone to post on ML about their 3rd party CI, but now I know better too :) | 18:05 | |
| anteaya | rloo: thank you | 18:06 |
| anteaya | anytime you have questions about third party ci just ask me | 18:06 |
| rloo | thank YOU anteaya for explaining/clarifying. | 18:06 |
| anteaya | it is much easier to just ask me or thingee and we can talk about it | 18:06 |
| anteaya | rloo: thanks for listening | 18:07 |
| anteaya | the third party ci group in my experience unfortunately has a mob mentality | 18:07 |
| anteaya | they will do anything anyone else does if they reach their goal | 18:07 |
| anteaya | they really aren't paying attention to the effect on the community | 18:07 |
| anteaya | again their are exceptions to this | 18:08 |
| anteaya | but they are the minority in my experience | 18:08 |
| rloo | anteaya: I guess we haven't (yet) seen that problem, so maybe that would be a good problem to have, but better to avoid it in the first place I think :D | 18:08 |
| devananda | anteaya: my experience is that many of the ironic driver authors are somewhat better community citizens than i've seen in other projects. | 18:08 |
| devananda | anteaya: not all of them, but on average we've had less issues than, say, neutron did at the point in time when 3rd party CI was a serious issue there | 18:09 |
| anteaya | rloo: exactly yes, I hope ironic never has that problem | 18:09 |
| anteaya | devananda: oh I'm so glad | 18:09 |
| * lucasagomes reads that conversation | 18:09 | |
| devananda | also, I think in a lot of ways that is because we've been able to learn from other projects' challenges :) | 18:09 |
| anteaya | devananda: yes I am glad to hear ironic third party ci folks are good community citizens | 18:10 |
| anteaya | devananda: that helps :) | 18:10 |
| *** harlowja has joined #openstack-ironic | 18:10 | |
| * mgould -> home: good night, all! | 18:10 | |
| NobodyCam | night mgould | 18:10 |
| * dtantsur calls it a day too | 18:10 | |
| *** mgould has quit IRC | 18:10 | |
| devananda | penick: not sure if you saw my ping ... you around today to chat? | 18:10 |
| *** baoli_ has joined #openstack-ironic | 18:10 | |
| *** chlong has quit IRC | 18:10 | |
| dtantsur | see you | 18:10 |
| NobodyCam | night dtantsur :) | 18:11 |
| devananda | g'night dtantsur | 18:11 |
| rloo | night dtantsur, mgoddard | 18:11 |
| *** dtantsur is now known as dtantsur|afk | 18:11 | |
| * sambetts is taking off too | 18:11 | |
| rloo | oops, sorry mgoddard, i meant mgould who's no longer here :) | 18:11 |
| rloo | bye sambetts | 18:11 |
| *** sambetts is now known as sambetts-afk | 18:11 | |
| *** cdearborn has quit IRC | 18:11 | |
| sambetts-afk | o/ rloo | 18:11 |
| *** paul-carlton1 has joined #openstack-ironic | 18:12 | |
| vdrok | good night everyone :) | 18:12 |
| mgoddard | rloo: no problem! | 18:12 |
| *** spandhe has joined #openstack-ironic | 18:13 | |
| *** baoli has quit IRC | 18:14 | |
| NobodyCam | night sambetts-afk | 18:14 |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Add documentation for proxies usage with IPA https://review.openstack.org/250878 | 18:14 |
| *** adam_g has quit IRC | 18:14 | |
| *** adam_g has joined #openstack-ironic | 18:14 | |
| *** adam_g has quit IRC | 18:15 | |
| *** adam_g has joined #openstack-ironic | 18:15 | |
| *** trown is now known as trown|lunch | 18:17 | |
| *** ndipanov has quit IRC | 18:17 | |
| *** baoli_ has quit IRC | 18:18 | |
| *** davidlenwell has quit IRC | 18:21 | |
| *** davideagnello has joined #openstack-ironic | 18:22 | |
| *** davidlenwell has joined #openstack-ironic | 18:23 | |
| *** Guest83268 has quit IRC | 18:23 | |
| *** krtaylor has quit IRC | 18:23 | |
| *** chlong has joined #openstack-ironic | 18:26 | |
| *** TheJulia has quit IRC | 18:26 | |
| penick | devananda: hey, I am. are you available at 1300h pacific | 18:26 |
| penick | ? | 18:27 |
| *** crinkle has quit IRC | 18:27 | |
| *** mgagne has joined #openstack-ironic | 18:28 | |
| *** mgagne is now known as Guest63453 | 18:28 | |
| *** crinkle has joined #openstack-ironic | 18:29 | |
| *** penick has quit IRC | 18:29 | |
| *** aginwala has joined #openstack-ironic | 18:31 | |
| *** penick has joined #openstack-ironic | 18:32 | |
| *** aweeks has quit IRC | 18:33 | |
| *** jcoufal has quit IRC | 18:33 | |
| *** ndipanov has joined #openstack-ironic | 18:33 | |
| *** paul-carlton1 has quit IRC | 18:33 | |
| *** jistr has joined #openstack-ironic | 18:37 | |
| *** Guest63453 has quit IRC | 18:38 | |
| *** Guest63453 has joined #openstack-ironic | 18:38 | |
| *** Guest63453 is now known as mgagne | 18:39 | |
| *** krtaylor has joined #openstack-ironic | 18:41 | |
| *** [1]cdearborn has quit IRC | 18:42 | |
| *** TheJulia has joined #openstack-ironic | 18:46 | |
| *** Marga_ has joined #openstack-ironic | 18:48 | |
| devananda | penick: I need to go in the office for a bit, but I can do that after 1300 | 18:50 |
| *** ipukha has quit IRC | 18:50 | |
| penick | ok, if that works for you i’ll drop you a line | 18:50 |
| devananda | great, thans | 18:51 |
| devananda | thanks | 18:51 |
| *** achanda has joined #openstack-ironic | 18:52 | |
| *** TheJulia has quit IRC | 18:55 | |
| *** penick has quit IRC | 18:56 | |
| *** athomas has quit IRC | 18:57 | |
| *** electrofelix has quit IRC | 18:59 | |
| *** aginwala has quit IRC | 19:02 | |
| openstackgerrit | Merged openstack/pyghmi: Add TLS support and TSM remote video (WIP) https://review.openstack.org/246596 | 19:04 |
| *** TheJulia has joined #openstack-ironic | 19:04 | |
| *** mkovacik has quit IRC | 19:06 | |
| *** aginwala has joined #openstack-ironic | 19:07 | |
| *** penick has joined #openstack-ironic | 19:08 | |
| *** trown|lunch is now known as trown | 19:12 | |
| *** Pavo has quit IRC | 19:13 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: [WIP] Fall back to old boot.ipxe behaviour if inc command is not found https://review.openstack.org/250745 | 19:14 |
| *** kozhukalov_ has joined #openstack-ironic | 19:16 | |
| *** Pavo has joined #openstack-ironic | 19:17 | |
| *** xek has quit IRC | 19:21 | |
| *** aginwala is now known as aginwala87 | 19:24 | |
| *** aginwala87 is now known as aginwala | 19:24 | |
| *** aginwala has quit IRC | 19:27 | |
| *** garthb has joined #openstack-ironic | 19:30 | |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic: Migrate to using keystoneauth Sessions https://review.openstack.org/236982 | 19:34 |
| *** sacharya has joined #openstack-ironic | 19:36 | |
| *** praneshp has joined #openstack-ironic | 19:36 | |
| openstackgerrit | Colleen Murphy proposed openstack/bifrost: Provide dhcp-range even for static inventory https://review.openstack.org/250079 | 19:37 |
| openstackgerrit | Merged openstack/ironic-inspector: Do not explicitly mention requirements.txt in tox.ini https://review.openstack.org/245754 | 19:40 |
| lucasagomes | folks I will call it a day | 19:41 |
| lucasagomes | have a great night | 19:41 |
| NobodyCam | have a good night lucasagomes | 19:41 |
| *** zer0c00l has joined #openstack-ironic | 19:42 | |
| *** lucasagomes is now known as lucas-dinner | 19:42 | |
| *** aginwala has joined #openstack-ironic | 19:43 | |
| *** garthb has quit IRC | 19:45 | |
| *** garthb_ has joined #openstack-ironic | 19:45 | |
| *** aarefiev22 has joined #openstack-ironic | 19:47 | |
| *** penick has quit IRC | 19:48 | |
| *** penick has joined #openstack-ironic | 19:50 | |
| NobodyCam | woo hoo new gaming laptop ordered | 19:51 |
| *** pas-ha has quit IRC | 19:55 | |
| *** garthb_ has quit IRC | 19:56 | |
| *** sinval-afk is now known as sinval | 19:56 | |
| *** thrash has joined #openstack-ironic | 19:56 | |
| *** thrash has quit IRC | 19:56 | |
| *** thrash has joined #openstack-ironic | 19:56 | |
| sinval | NobodyCam: \o/ | 19:56 |
| *** garthb has joined #openstack-ironic | 19:57 | |
| NobodyCam | :) I get a new one about ever 5 years or so :) | 19:57 |
| *** Pavo has quit IRC | 19:58 | |
| sinval | NobodyCam: cool | 19:59 |
| *** rajathagasthya has joined #openstack-ironic | 19:59 | |
| *** penick has quit IRC | 20:03 | |
| *** [1]cdearborn has joined #openstack-ironic | 20:03 | |
| *** sacharya has quit IRC | 20:05 | |
| *** rajathagasthya has quit IRC | 20:05 | |
| *** penick has joined #openstack-ironic | 20:08 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Switch to Ansible 2.0 https://review.openstack.org/241330 | 20:19 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Allow user to set DIB_RELEASE for disk images https://review.openstack.org/251526 | 20:19 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Change DIB role to default to Debian jessie https://review.openstack.org/251527 | 20:19 |
| *** Marga_ has quit IRC | 20:19 | |
| openstackgerrit | Merged openstack/python-ironicclient: Add missing translation markers https://review.openstack.org/246875 | 20:21 |
| *** jistr has quit IRC | 20:27 | |
| TheJulia | SpamapS: you might want to checkout the revs ^^^ which should address a concern you had expressed. | 20:28 |
| *** garthb has quit IRC | 20:30 | |
| *** achanda has quit IRC | 20:31 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Set variables for Ubuntu 15.10 https://review.openstack.org/243827 | 20:37 |
| *** aginwala has quit IRC | 20:39 | |
| *** aginwala has joined #openstack-ironic | 20:42 | |
| *** alexpilo_ has joined #openstack-ironic | 20:44 | |
| *** alexpilotti has quit IRC | 20:45 | |
| *** alexpilo_ has quit IRC | 20:45 | |
| *** alexpilotti has joined #openstack-ironic | 20:45 | |
| *** alexpilotti has quit IRC | 20:46 | |
| *** alexpilotti has joined #openstack-ironic | 20:46 | |
| *** alexpilotti has quit IRC | 20:47 | |
| *** alexpilo_ has joined #openstack-ironic | 20:47 | |
| *** alexpilo_ has quit IRC | 20:47 | |
| *** Marga_ has joined #openstack-ironic | 20:50 | |
| *** dims_ has joined #openstack-ironic | 20:53 | |
| *** dims has quit IRC | 20:56 | |
| *** Pavo has joined #openstack-ironic | 20:57 | |
| *** aginwala has quit IRC | 21:02 | |
| *** dims_ has quit IRC | 21:03 | |
| *** Piet has quit IRC | 21:04 | |
| *** Marga_ has quit IRC | 21:05 | |
| penick | devananda: hey hey | 21:06 |
| *** Marga_ has joined #openstack-ironic | 21:06 | |
| *** rajathagasthya has joined #openstack-ironic | 21:06 | |
| devananda | penick: yo | 21:06 |
| penick | What’s shakin’ sir? | 21:06 |
| devananda | wanna talk about the claims / search api | 21:07 |
| devananda | and scheduling things | 21:07 |
| penick | Absolutely | 21:07 |
| penick | what’re your thoughts? | 21:07 |
| *** kozhukalov_ has quit IRC | 21:07 | |
| *** dprince has quit IRC | 21:07 | |
| devananda | I neither want to reimplement nova-scheduler in totality, nor want to remain stuck with the current limitation to a single nova-compute host | 21:08 |
| penick | I’m sensing a “but” | 21:09 |
| devananda | heh | 21:09 |
| devananda | your objection on the spec, IIUC, is baslically "don't take away my pluggable python scheduler filter API" | 21:09 |
| *** Marga_ has quit IRC | 21:10 | |
| devananda | however, the Nova project doesn't, as far as I can tell, have any interest in solving for our problem domain. they would prefer baremetal-related problems be solved in the baremetal project (including scheduling decisions that do not relate to VMs) | 21:11 |
| penick | It’s partially that, but moreso it’s when I take a step back and look at the situation. We have a system that pulls data from one table in a DB, and plugs it into another table in a DB. That’s inefficient and it’s hurting things. The solutions thus far seem predicated on the assertion that Nova won’t allow anything non-VM any deeper. So the Ironic integration is designed to be an absolute minimum of touch points | 21:12 |
| *** kozhukalov_ has joined #openstack-ironic | 21:13 | |
| *** aginwala has joined #openstack-ironic | 21:13 | |
| penick | Which is what leads us to these pretty inefficient designs. But, if there was an Ironic specific ComputeNodes filter, it would gracefully plug in to the existing scheduling system without any other advanced changes | 21:13 |
| devananda | "plugs it into another table in a DB" -- aiui, the nova resource tracker is an in-memory object, not a db table | 21:13 |
| penick | True | 21:14 |
| devananda | so are you referring to something else there? or am I mistaken? | 21:14 |
| *** dims has joined #openstack-ironic | 21:14 | |
| devananda | actually, I think a large problem with nova's scheduler is that fetching a lot of data from a database into a python process and then filtering it *there* is a very inefficient design | 21:15 |
| devananda | I'd like to make ironic's scheduling much faster by actually leveraging the database for what it's good at -- searching through data | 21:16 |
| penick | One of the biggest inefficiencies was that the scheduler used to fetch data from a DB several times within a call. but that’s fixed in liberty. The first filter will hit the DB, and the remaining ones should just iterate over that | 21:16 |
| penick | How does that plug in with the rest of the filters we’re creating to represent the rest of our infrastructure? As we build rack/network/power anti-affinity, how does that mesh with Ironic doing its own Scheduling? | 21:17 |
| devananda | penick: I don't know yet | 21:18 |
| devananda | penick: that locality/failure-domain data must be stored in some format within ironic, regardless of where the scheduling happens | 21:19 |
| devananda | penick: could those scheduler hints / filters be represented within the query/claim API we've proposed? | 21:19 |
| *** aginwala has quit IRC | 21:19 | |
| *** Marga_ has joined #openstack-ironic | 21:19 | |
| penick | they could, for the low price of my team writing all of it twice.. :) | 21:20 |
| penick | One set for VM one for BM | 21:20 |
| devananda | penick: how about ironic virt driver converting the nova request into the appropriate query to ironic? | 21:20 |
| devananda | someone write that code once and everyone uses it | 21:21 |
| penick | Is this really something that needs to be written in Ironic, or can node selection just happen from a Ironic-Specific filter? | 21:21 |
| devananda | penick: if nova-scheduler is used to select an ironic node _at_all_, how does it know which nova-compute host to send the request to? | 21:21 |
| devananda | the primary reason for this work is to address the current limitation of running only a single nova-compute host for Ironic | 21:22 |
| devananda | oddly, the spec doesn't mention that in the problem statement | 21:22 |
| devananda | but this spec has emerged from about a year of discussion with the nova team trying to find ways to solve that | 21:23 |
| penick | If the scheduling moves to the scheduler, then the compute node selection could be a simple load balancing algorithm. Find one that isn’t busy, give it a job to do | 21:23 |
| penick | and integrate that with the claims process | 21:23 |
| devananda | penick: https://review.openstack.org/#/c/194453/ | 21:23 |
| devananda | penick: how does each compute host report resource availability to the scheduler? | 21:23 |
| devananda | which AVAILABLE nodes does each compute host report? what about cells support? | 21:24 |
| lifeless | penick: btw, you have mail | 21:24 |
| penick | Which type of resource availability? As in, how many jobs the compute host is in the midst of. Or, the sum of the capabilities of all nodes managed by that compute node? | 21:25 |
| penick | lifeless: i’ll check, danke! | 21:25 |
| devananda | penick: "all nodes managed by that compute node" << this is the list of all nodes in AVAILABLE state within Ironic, and is identical on every compute node | 21:26 |
| openstackgerrit | Summer Barringer proposed openstack/bifrost: Adjusted line length to 79 columns. https://review.openstack.org/251544 | 21:26 |
| penick | If we have the nova scheduler more able to directly access relevant scheduling data, why have the compute node report anything? That’s first used as part of the computenodes filter, yeah? If there’s an IronicComputeNode filter that knows to query the Ironic DB directly to create the in-memory object | 21:26 |
| penick | devananda: I think the problem is the way I envisiion it - having a compute node report the capabilities of the hosts it manages is fundamentally wrong. It makes sense for VM. But doing that for baremetal means we’re just trying to cram Ironic into a VM shaped mold. And with a few changes we could satisfy both systems. | 21:28 |
| penick | With VM a compute node reports its capabilities up because a: it’s a 1:1 relationshuip and b: it changes based on utilization, current overcommit values, etc | 21:29 |
| penick | With baremetal the capabilities of a machine are not as ephemeral. It’s either in use, or it’s not in use. Only in the event of upgrades is there a state change | 21:29 |
| penick | So, given that, if we accept that Ironic will integrate with the Nova scheduler in a different way, we can remove the assumption that the compute node is anything but a dumb worker that performs a task on a claimed host | 21:30 |
| devananda | penick: nova team didn't accept that proposal when I made it a few summits back | 21:31 |
| penick | I feel like they might be more open minded now | 21:31 |
| devananda | it essentially requires gutting the nova resource tracker and compute host logic, and significantly changing the scheduler so it integrates with an external (from nova's POV) system | 21:31 |
| devananda | what we're proposing here doesn't prevent that from being done later on | 21:32 |
| devananda | but it allows us to move forward with something while nova team continues to refactor the RT and scheduler | 21:32 |
| penick | It doesn’t, but it does mean that I have to be out-of-tree until it does, which fundamentally is something we really really really want to avoid | 21:33 |
| penick | s/we/I/ | 21:33 |
| devananda | I don't want (you) to be out of tree (either) | 21:33 |
| penick | In our concurrency scale test we up and disabled RT, since it was just in the way | 21:35 |
| devananda | heh | 21:35 |
| devananda | why does that not surprise me :) | 21:35 |
| *** Pavo has quit IRC | 21:35 | |
| penick | I don’t think the scheduler re-factor should be a blocker — quite the opposite in fact. If someone’s going to re-factor the scheduler I want it done with an Ironic driver at the beginning of the chain | 21:35 |
| devananda | penick: you talked with jaypipes yet? | 21:36 |
| penick | On the scheduler stuff, not really | 21:36 |
| devananda | you might want to | 21:36 |
| penick | I should go chase him down this week and see if he’s got time | 21:36 |
| *** aginwala has joined #openstack-ironic | 21:37 | |
| penick | So that seems probably the best step from this point. -If- Nova agrees that, yes, Ironic should be its own filter and not some glued-on thing — would you agree that this is the best way forward? | 21:40 |
| *** Piet has joined #openstack-ironic | 21:41 | |
| devananda | Ironic requires much more than just a Nova Filter plugin | 21:41 |
| devananda | I do not agree that that will solve the fundamental problem here | 21:41 |
| *** dims_ has joined #openstack-ironic | 21:42 | |
| *** baoli has joined #openstack-ironic | 21:43 | |
| *** baoli has quit IRC | 21:43 | |
| *** dims has quit IRC | 21:43 | |
| penick | Sure, the three main problems I see are 1: Create ironic filter, have other filters accept ironic-y data that comes through in the object 2: claims API to reserve a node as a part of the scheduling process. 3: cook up a mechanism to select a compute node to service a request when a node’s been selected | 21:46 |
| penick | 4: if it’s Ironic, leave RT in the box like a christmas fruitcake. | 21:46 |
| *** caiobo_ has quit IRC | 21:47 | |
| *** aginwala has quit IRC | 21:49 | |
| devananda | afk a bit | 21:50 |
| devananda | tiny emergency involving an orange, a knife, and a finger | 21:51 |
| penick | ah, yeah, no problem | 21:51 |
| openstackgerrit | Merged openstack/python-ironicclient: Refactoring and removing duplicate code of "base.Manager" heirs https://review.openstack.org/246692 | 21:51 |
| *** baoli has joined #openstack-ironic | 21:52 | |
| *** penick has quit IRC | 21:56 | |
| *** penick has joined #openstack-ironic | 21:59 | |
| *** aginwala has joined #openstack-ironic | 21:59 | |
| SpamapS | TheJulia: ACK! THanks! | 22:05 |
| *** Marga__ has joined #openstack-ironic | 22:05 | |
| *** aarefiev22 has quit IRC | 22:05 | |
| *** Marga_ has quit IRC | 22:06 | |
| *** Pavo has joined #openstack-ironic | 22:07 | |
| *** Marga__ has quit IRC | 22:09 | |
| *** praneshp has quit IRC | 22:11 | |
| *** trown is now known as trown|outttypeww | 22:16 | |
| *** robcresswell has quit IRC | 22:17 | |
| *** robcresswell has joined #openstack-ironic | 22:18 | |
| NobodyCam | devananda: hope your ok? | 22:20 |
| NobodyCam | hey SpamapS :) | 22:20 |
| *** alexpilotti has joined #openstack-ironic | 22:21 | |
| mariojv | thanks for pointing out where the cache is being cleared out vdrok, must have missed that | 22:24 |
| *** kozhukalov_ has quit IRC | 22:39 | |
| *** rajathagasthya has quit IRC | 22:39 | |
| *** rajathagasthya has joined #openstack-ironic | 22:41 | |
| *** kkoski has joined #openstack-ironic | 22:43 | |
| *** penick has quit IRC | 22:52 | |
| krotscheck | Argh, I keep forgetting the meeting | 22:54 |
| *** praneshp has joined #openstack-ironic | 22:55 | |
| NobodyCam | krotscheck: good thing its recorded :p | 22:56 |
| *** Marga_ has joined #openstack-ironic | 22:56 | |
| krotscheck | NobodyCam: I know, I'm reading the minutes | 22:56 |
| NobodyCam | :) | 22:56 |
| *** Marga_ has quit IRC | 22:57 | |
| *** Marga_ has joined #openstack-ironic | 22:57 | |
| *** jobewan has quit IRC | 22:57 | |
| *** praneshp_ has joined #openstack-ironic | 22:58 | |
| openstackgerrit | William Stevenson proposed openstack/python-ironicclient: Updates supporting ironic-neutron integration https://review.openstack.org/206144 | 22:58 |
| *** praneshp has quit IRC | 23:00 | |
| *** praneshp_ is now known as praneshp | 23:00 | |
| *** Marga_ has quit IRC | 23:01 | |
| *** rajathagasthya has quit IRC | 23:02 | |
| *** dims has joined #openstack-ironic | 23:02 | |
| *** dims_ has quit IRC | 23:03 | |
| *** rajathagasthya has joined #openstack-ironic | 23:03 | |
| *** Marga_ has joined #openstack-ironic | 23:04 | |
| *** aginwala_ has joined #openstack-ironic | 23:06 | |
| *** aginwala has quit IRC | 23:10 | |
| *** caiobo_ has joined #openstack-ironic | 23:16 | |
| *** thrash is now known as thrash|g0ne | 23:21 | |
| *** rpioso has quit IRC | 23:24 | |
| *** alexpilotti has quit IRC | 23:29 | |
| *** barra204 has quit IRC | 23:38 | |
| *** shakamunyi has joined #openstack-ironic | 23:38 | |
| *** jerrygb has quit IRC | 23:39 | |
| *** jerrygb has joined #openstack-ironic | 23:39 | |
| *** [1]cdearborn has quit IRC | 23:40 | |
| *** HimanshuG has joined #openstack-ironic | 23:43 | |
| *** HimanshuG has left #openstack-ironic | 23:45 | |
| *** smoriya_ has joined #openstack-ironic | 23:48 | |
| *** alexpilotti has joined #openstack-ironic | 23:49 | |
| *** HimanshuGarg has joined #openstack-ironic | 23:51 | |
| *** penick has joined #openstack-ironic | 23:54 | |
| *** alexpilotti has quit IRC | 23:54 | |
| *** HimanshuGarg has quit IRC | 23:55 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!