Wednesday, 2017-10-25

*** kumarmn has quit IRC00:31
*** mriedem has quit IRC00:56
zanebharlowja: so say a hypervisor dies. your HA will fail over. but you also need to replace the dead VM, right? that's the OpenStack native part00:57
harlowjagotcha00:57
zaneband I guarantee that you will run into all sorts of interesting problems that effectively prevent users from doing this kind of cloud-like thing on OpenStack00:59
harlowjayup00:59
harlowjaopenstack to low level00:59
zanebe.g. there's no way to get notifications that a hypervisor has died, because Nova doesn't believe that's a thing that users will ever need00:59
harlowjasure, that to00:59
zanebyou'll have to put your cloud credentials on a VM somewhere because Keystone hasn't implemented application credentials yet (although mordred is now working on it)01:00
harlowjazaneb though how does heat do it then? or how did it simulate that01:00
harlowjaceliometer back channel...01:00
zanebharlowja: there's _some_ stuff you can do with ceilometer01:01
zanebbut not enough01:01
harlowjaya, guess we forgot to define01:01
harlowja'OpenStack native ' (from the start)01:01
harlowjalol01:01
harlowjaoops01:01
harlowjasorry bout that, ha01:01
zanebharlowja: e.g. I wrote this http://git.openstack.org/cgit/openstack/heat-templates/tree/hot/autohealing/autohealing_server.yaml but to use it you need to enable some config in ceilometer that ~nobody enables01:02
zanebI don't care if you use Heat or not01:02
harlowjaya, i get it01:02
* harlowja thinks we need to make diagram like https://raw.githubusercontent.com/cncf/landscape/master/landscape/CloudNativeLandscape_v0.9.7.jpg lol01:03
zanebbut I would love for someone to build the demo of what SamYaple described, and then come back with a list of all the things that you can't actually do in OpenStack and then pull the fire alarm01:03
harlowjamore diagrams, hahahaha01:03
harlowjazaneb or just admit that perhaps people don't want to do those things anymore in openstack (if they ever did in the first place)?01:04
*** kumarmn has joined #openstack-tc01:05
zanebharlowja: that'd be fine too, we could change the mission statement and make it clear to the world that Nova/Glance/Neutron/Cinder are all you're ever gonna get and if you want to build cloud-like things on top then you're going to be writing your own thing that does a lot of polling (like nodepool, maybe?)01:06
zaneband I can go work on something else01:06
harlowjaya, how's our outreach programs going :-P01:06
zanebthat diagram is interesting. apparently CNCF would like people to think that the only thing OpenStack is good for is bare-metal provisioning01:09
harlowjaya, i raised that on openstack-dev a few weeks ago01:10
zanebwhich we don't even do well01:10
harlowjabeing, like ummm, does anyone know who to contact to fix that, lol01:10
harlowjahttp://lists.openstack.org/pipermail/openstack-dev/2017-September/122203.html lol01:10
fungiildikov has been trying, i think. or maybe it was lsell01:11
harlowjacool01:11
fungiharlowja: aiui, that diagram is the checklist of projects the cncf wants to recruit to help boost their image01:11
harlowjaya, i know01:11
harlowja(not hard to tell that, ha)01:11
harlowja*all the boxes*01:12
harlowjahttps://lists.cncf.io/pipermail/cncf-toc/2017-September/001197.html i thought was funny, ha01:12
*** kumarmn has quit IRC01:12
fungithey apparently just convinced tuf and notary to join01:12
harlowjaya, in some fantasy world of mine, CNCF and openstack foundation would be the same thing01:13
fungiit's an interesting tactic, taking existing reasonably successful projects and convincing them to become part of the cncf01:13
harlowjaya, slightly different then the 'built in openstack' only problem01:14
*** mriedem has joined #openstack-tc01:15
fungioh, and i guess we've officially been in office hour for 16 minutes now01:16
*** Qiming has quit IRC01:20
*** persia has quit IRC01:21
*** persia has joined #openstack-tc01:22
harlowjaoh nice, ha01:22
fungiand everybody clams up once i point that out ;)01:23
zanebyeah, way to kill the discussion fungi ;)01:23
* fungi is used to it01:23
harlowjanah, talking to my manager01:26
harlowjanot clammy01:26
fungiharlowja: not as much "built in openstack" but rather that we have projects coming to us asking to join, to the point that we don't really go looking for projects to talk into becoming an openstack project01:26
fungimaybe they're just looking to reach critical mass to the point where projects will seek them out rather than having to go about it the other way around01:27
*** Qiming has joined #openstack-tc01:28
*** mriedem has quit IRC01:28
EmilienMfungi: any examples we can discuss tonight?01:29
fungiEmilienM: zaneb and harlowja's conversation about a self-healing "hello world" application using openstack native solutions is an interesting thought exercise, at least01:31
fungiEmilienM: not sure what sort of examples you're looking for01:32
EmilienMfungi: I wasn't sure if you were refering to recent applications request01:32
EmilienMI agree it's an interesting exercise01:32
fungioh, nope. just all of them really01:32
EmilienMok01:33
fungiwe get lots of great projects applying to join01:33
*** Qiming has quit IRC01:33
fungiSamYaple just pushed up one for loci01:33
EmilienMyeah, I saw that today, this is really good.01:33
* fungi feels bad, has been too busy to read through it properly yet01:34
*** Qiming has joined #openstack-tc01:35
*** Qiming has quit IRC01:39
SamYaplefungi: its cool, i give you permission to skip it if you +1 it in return01:40
*** Qiming has joined #openstack-tc01:40
fungibwahahaha01:40
EmilienMhehe01:42
EmilienMfungi: he didn't have to do the same with me, I gave it before :P01:42
*** Qiming has quit IRC02:22
*** Qiming has joined #openstack-tc02:52
*** Qiming has quit IRC02:56
openstackgerritDai Dang Van proposed openstack/governance master: Add policy artifacts for senlin  https://review.openstack.org/51490804:37
*** alex_xu has quit IRC04:56
*** alex_xu has joined #openstack-tc04:59
*** alex_xu has quit IRC06:09
*** alex_xu has joined #openstack-tc06:15
ildikovfungi: that was lsell in this case :)06:20
*** gcb has joined #openstack-tc07:00
ttxroster updated at https://www.openstack.org/foundation/tech-committee/ -- you might want to check/update your bios and affiliations on OpenStackID if not correct07:07
*** gcb has quit IRC07:57
ttxtc-members if you have people you would like to nominate for a contributor award in Sydney, you probably still have a couple of hours to do so07:59
*** sdague has joined #openstack-tc09:20
*** cdent has joined #openstack-tc09:57
cdentthat dinner location looks nice10:12
*** dtantsur|afk is now known as dtantsur11:01
*** Guest3211 has quit IRC11:59
*** amrith has joined #openstack-tc12:14
*** amrith is now known as Guest723012:15
*** kumarmn has joined #openstack-tc12:15
*** kumarmn has quit IRC12:20
*** rosmaita has joined #openstack-tc12:22
*** Guest7230 has quit IRC12:31
*** amrith_ has joined #openstack-tc12:35
*** amrith_ is now known as Guest3070712:35
*** mriedem has joined #openstack-tc12:50
*** kumarmn has joined #openstack-tc13:14
*** kumarmn has quit IRC13:18
*** kumarmn has joined #openstack-tc13:33
ttxOh and btw if you see anything we should really be discussing in Sydney at the Forum, there should be extra Forum slots set aside to cover for last-minute needs that we can reuse13:36
cdentttx: flaper87 suggested perhaps a followup on the topics that arose from the term limits conversation13:40
mugsieI think that might be a good candidate for a f2f conversation13:44
flaper87yeah, I would like to pursue that conversation further. I think it'll provide good food for thoughts and it'll help us revisit some (tribal?) knowledge13:47
flaper87this is not to say I agree/disagree either way but I wouldn't mind having that conversation13:47
flaper87I think EmilienM is planning to send an email out13:47
EmilienMflaper87: I'll, when you do my breakfast.13:48
flaper87also, I won't be at the TC+Board meeting so, let's please not discuss this there :D13:48
EmilienMmilk + juice please13:48
flaper87EmilienM: gimme and address, I'll call uber eats13:48
flaper87or dunkey eats, depending where you are13:48
TheJulialol13:49
EmilienMthere is no über on the Island that I live13:49
flaper87EmilienM: I'm sure I can hire a donkey to get you food.13:49
flaper87EmilienM: worst case scenario, you eat the donkey13:50
* flaper87 stfu and stops spamming the channel13:50
* EmilienM stfu as well and prepares an email13:50
openstackgerritThierry Carrez proposed openstack/governance master: Add Storyboard Migration to Rocky  https://review.openstack.org/51387513:51
*** hongbin has joined #openstack-tc14:08
*** Guest30707 is now known as amrith14:10
*** david-lyle has quit IRC14:44
*** david-lyle has joined #openstack-tc14:53
pabelangerttx: roster looks good, but noticed linkedin URLs are not correct, they are invalid URLs14:54
ttxpabelanger: ok, will communicate issue up15:01
ttxpabelanger: they work for me15:02
ttx(open a new tab on https://twitter.com/pabelanger)15:02
ttxoh, linked in15:03
ttxignore me15:03
cdentI suppose I should tidy up my bio if it is going to be so…available15:12
EmilienMcdent: just write "hi i'm cdent and here for you"15:47
cdentand you and you and you15:47
smcginnisHah15:52
flaper87cdent: and me?15:57
cdentand you!15:57
flaper87(emoji with heartened eyes)15:57
cdent😍15:58
flaper87es, it is cheating..but heck you want to get a space, right?15:59
flaper87arg, bad paste15:59
flaper87😍15:59
flaper87that one15:59
*** Rockyg has joined #openstack-tc16:15
EmilienMhow do you do that16:17
EmilienMwe need a doc16:17
cdentOr a resolution to not allow emoji here16:17
EmilienMRollcall +116:18
cdentcommunalism finally comes to openstack: if not everyone can do it, then no one can do it!16:19
EmilienMwell, you didn't tell me how to do it :P16:21
cdent💩❤️❤️👾👾🤖🎃👺💩👊👊16:22
cdenti have one of those useless touchbars16:22
cdentthat ^^ is pretty much all they are good for16:22
EmilienMit hurts my eyes16:23
smcginnisWe have become slack.16:24
Rockygare you sure we're not just slackers?16:25
mugsiewell, that is a given :)16:28
mugsieespecially if we make our touchbars have emjois :)16:28
* dtroyer wonders if he can resurrect his old slackersatowrk.com domain16:49
openstackgerritKendall Nelson proposed openstack/governance master: Add Storyboard Migration to Rocky  https://review.openstack.org/51387516:56
*** dtantsur is now known as dtantsur|afk17:01
ttxtc-members: for a board+TC+UC meeting in Dublin, would you prefer Sunday, Monday or Friday ?17:04
cdentsunday17:04
cmurphyif the next PTG is anything like the last one I will be busy monday-friday17:05
smcginnisAny would likely work for me.17:05
cdentcmurphy: ditto17:05
smcginniscmurphy: Good point, Monday might not be the best.17:05
dhellmannsunday17:05
dtroyerI'd prefer Sunday17:05
* dhellmann wonders why we keep having this come up as a question, since we always want it to be a separate day17:05
smcginnis:)17:06
ttxdhellmann: cost mostly17:06
* ttx wonders if we really need to meet, too17:06
ttxSydney feels just around the corner from Denver to me17:07
EmilienMsunday17:07
cdentit’s fair to want to limit costs, but if we want to meet, it needs to be dedicated time, otherwise we can’t do the PTG justice17:07
dhellmannmaybe skip the meetings at the ptg and only meet at summits?17:07
ttxdhellmann: right, my thought exactly17:07
cdentI _like_ having the meetings, but they are perhaps not _necessary_17:07
EmilienMsunday with guinness as post-meeting17:08
cdentI feel, however, that if we don’t have them, we miss a lot of opportunity for brainshare17:08
smcginniscdent: Same here.17:08
pabelangerttx: sunday17:08
cdentanothr way to put it: if we don’t have the meetings then my interaction with the board will drop to pretty much nothing, and that seems bad17:09
smcginnisI agree. It's soon after Denver, but I do think the time together face to face is useful.17:10
ttxok, will communicate our (usual) preference17:10
ttxwe still don't have an agenda for Sydney meeting :)17:11
dhellmanncdent : yeah; I think I would prefer to go ahead and have them, too. These groups meet so infrequently as a body, and I think it's useful. We (all) need to be very active about managing the agenda, though, to ensure that.17:11
cdentdhellmann: yes. ttx: as (I think) I said before: I can have a very small bit prepared based on the ptg questions about “satisfaction”17:12
ttxcdent: ok17:13
pabelangerIs flattening of git namespaces something that could be talked about? I admit, I am unsure the current status of that17:13
ttxwill add it to the wiki page if you don't beat me to it17:13
cdentttx: if you could, that would be great please and thank you, I’m holding a lot of code state atm17:14
ttxpabelanger: I expect it will be discussed as a consequence of the bucketing, yes17:14
dhellmannpabelanger : do the board and uc have input into that in some way? or interest?17:14
dhellmannit probably fits into the discussion of new project areas, so it might come up there17:14
pabelangerdhellmann: I am not sure to be honest17:15
ttxdhellmann: yes it may come up as an aside17:15
dhellmannpabelanger : ok, I was mostly looking for your perspective on why that might be an interesting topic (not suggesting that it isn't)17:15
* ttx dinners17:16
pabelangerdhellmann: Yah, I find it an interesting topic, mostly because of the infra changes that will be needed. However, I wasn't sure if that given the green light yet or not from board. I do recall reading some slights at PTG about it, and wasn't sure what the follow up was.17:19
*** dmsimard is now known as dmsimard|afk17:23
mugsieis there any info on that bucketing anywhere? I seem to remember reading a high level overview powerpoint, but I can't find it17:25
cdentmugsie: you mean the “new projects that are not openstack but in the foundation” thing? or something else?17:26
mugsieyeah17:26
cdentThat powerpoint is all I’ve seen17:27
cdentIt feels, however, like the plan has advanced from there somewhat, but I don’t know of any documents17:28
cdentAn email has been promised, prior to summit, with more information17:28
cdentin part to forestall any “how will this impact _me_” concerns17:28
pabelangerhttps://docs.google.com/presentation/d/1qQ1oCjhEE3bjtjkPydGhyeR6vFyuyFeOhbwc-WueddI/edit#slide=id.g253ac459b6_0_20817:30
pabelangerYah, I should have used 'buckets' over flaten namespaces17:30
cdentthat’s different from the one I was talking about, but seems more likely to be the one that mugsie was thinking about17:31
cdentI was thinking of17:31
cdenthttps://docs.google.com/presentation/d/1vhBRo7jLG0fmzktdnG1EhHwE5L7eYdIFkuxYXmGZ_Hg/edit#slide=id.g2634b3cd1a_0_1417:32
pabelangercdent: I don't think I have seen this one yet.  Will read shortly17:34
mugsieyeah - I had just seen the first one17:40
TheJuliattx: I prefer Sunday as well.17:43
*** Rockyg has quit IRC17:49
dhellmannpabelanger : sure. I thought it was basically an implementation detail but maybe there's more to it.17:51
dhellmannpabelanger : ah, the buckets thing is definitely something that will come up.17:51
* dhellmann has finally caught up with scrollback17:52
*** kumarmn_ has joined #openstack-tc18:32
*** kumarmn has quit IRC18:36
*** cdent has quit IRC18:38
*** marst has joined #openstack-tc19:32
*** rosmaita has quit IRC21:00
*** dmsimard|afk is now known as dmsimard21:08
fungittx: sunday before ptg in dublin has my vote as well, for all the usual reasons21:24
fungipabelanger: i have some ideas on ways we can make significant perceptual impact hiding namespaces in prominent places now with relatively minimal infra effort (i.e. not having to pull the trigger on repo renames in gerrit/jobs/et cetera until later)21:26
fungithough it also furthers my not-so-secret agenda of getting the infra team out of being particularly responsible for gh integration21:27
*** rosmaita has joined #openstack-tc21:33
*** sdague has quit IRC21:56
*** marst has quit IRC21:58
*** kumarmn_ has quit IRC22:03
*** kumarmn has joined #openstack-tc22:10
*** kumarmn has quit IRC22:15
*** hongbin has quit IRC23:05
*** kumarmn has joined #openstack-tc23:32
*** kumarmn has quit IRC23:37
*** mriedem is now known as mriedem_away23:44
*** mriedem_away has quit IRC23:44
*** kumarmn has joined #openstack-tc23:45

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!