| *** jamesmcarthur has joined #openstack-tc | 00:10 | |
| *** e0ne has joined #openstack-tc | 00:20 | |
| *** jamesmcarthur has quit IRC | 00:29 | |
| *** jamesmcarthur has joined #openstack-tc | 00:49 | |
| *** jamesmcarthur has quit IRC | 01:10 | |
| *** jamesmcarthur has joined #openstack-tc | 01:12 | |
| *** e0ne has quit IRC | 01:20 | |
| *** jamesmcarthur has quit IRC | 01:22 | |
| *** e0ne has joined #openstack-tc | 01:27 | |
| *** jamesmcarthur has joined #openstack-tc | 01:33 | |
| *** ricolin has joined #openstack-tc | 01:33 | |
| *** mriedem_away is now known as mriedem | 01:55 | |
| *** jamesmcarthur has quit IRC | 02:01 | |
| *** diablo_rojo has quit IRC | 02:02 | |
| *** mriedem has quit IRC | 02:38 | |
| *** e0ne has quit IRC | 02:40 | |
| *** e0ne has joined #openstack-tc | 02:48 | |
| *** jamesmcarthur has joined #openstack-tc | 02:52 | |
| *** openstackgerrit has joined #openstack-tc | 02:52 | |
| openstackgerrit | Zhipeng Huang proposed openstack/governance master: Add long term goal proposal to community wide goals https://review.openstack.org/602799 | 02:52 |
|---|---|---|
| *** jamesmcarthur has quit IRC | 02:57 | |
| *** lbragstad has quit IRC | 03:04 | |
| *** e0ne has quit IRC | 03:05 | |
| *** lbragstad has joined #openstack-tc | 03:12 | |
| *** jamesmcarthur has joined #openstack-tc | 03:13 | |
| *** ricolin has quit IRC | 03:13 | |
| *** ricolin has joined #openstack-tc | 03:15 | |
| *** jamesmcarthur has quit IRC | 03:17 | |
| *** jamesmcarthur has joined #openstack-tc | 03:33 | |
| *** jamesmcarthur has quit IRC | 03:37 | |
| *** lbragstad has quit IRC | 03:48 | |
| *** jamesmcarthur has joined #openstack-tc | 03:54 | |
| *** jamesmcarthur has quit IRC | 03:58 | |
| *** ricolin has quit IRC | 04:03 | |
| *** diablo_rojo has joined #openstack-tc | 04:14 | |
| *** jamesmcarthur has joined #openstack-tc | 04:14 | |
| *** jamesmcarthur has quit IRC | 04:19 | |
| *** diablo_rojo has quit IRC | 04:48 | |
| *** jaosorior has joined #openstack-tc | 04:54 | |
| *** jamesmcarthur has joined #openstack-tc | 04:55 | |
| *** jamesmcarthur has quit IRC | 05:00 | |
| *** diablo_rojo has joined #openstack-tc | 05:10 | |
| *** jamesmcarthur has joined #openstack-tc | 05:22 | |
| *** jamesmcarthur has quit IRC | 05:27 | |
| *** diablo_rojo has quit IRC | 05:44 | |
| *** jamesmcarthur has joined #openstack-tc | 06:04 | |
| *** jamesmcarthur has quit IRC | 06:08 | |
| *** jamesmcarthur has joined #openstack-tc | 06:25 | |
| *** jamesmcarthur has quit IRC | 06:29 | |
| *** jamesmcarthur has joined #openstack-tc | 06:46 | |
| *** jamesmcarthur has quit IRC | 06:50 | |
| *** jamesmcarthur has joined #openstack-tc | 07:27 | |
| *** jamesmcarthur has quit IRC | 07:31 | |
| *** jpich has joined #openstack-tc | 07:40 | |
| evrardjp | o/ | 07:51 |
| *** persia has quit IRC | 08:06 | |
| *** persia has joined #openstack-tc | 08:07 | |
| *** tosky has joined #openstack-tc | 08:16 | |
| ttx | o/ | 08:38 |
| * ttx emerges from the jetlag fog | 08:38 | |
| *** cdent has joined #openstack-tc | 08:46 | |
| *** lbragstad has joined #openstack-tc | 08:59 | |
| evrardjp | emerge can take a while (gentoo-ists will hate me there) | 09:04 |
| cmurphy | o/ i think it's office hours time (cc tc-members) | 09:04 |
| cdent | oh hair | 09:05 |
| cdent | hai! | 09:05 |
| evrardjp | ohai cdent | 09:05 |
| cdent | I got home and slept for 18 hours | 09:08 |
| ttx | o/ | 09:08 |
| cmurphy | cdent: I did the same | 09:08 |
| cmurphy | we'll see in a few hours if it takes or if I crash | 09:08 |
| cdent | my history suggests crash | 09:10 |
| cmurphy | likely | 09:11 |
| lbragstad | I was thinking about the service dependency conversation from Friday afternoon - would the constellations work make the justifications for operators more clear? | 09:11 |
| ttx | lbragstad: potentially yes | 09:12 |
| * cmurphy does timezone arithmetic, raises eyebrow at lbragstad | 09:12 | |
| lbragstad | i don't even have jet lag excuses for being up this early ;) | 09:13 |
| ttx | The conclusion of that discussion to me was that we need to say that dependencies are costly, so you should only depend on something if you can deliver user-facing gain (or such development-facilitating gain that the user would indeirectly benefit from it), and document that clearly | 09:13 |
| lbragstad | yeah - being explicit about it made sense, (penick and jroll did a good job of communicating that) | 09:14 |
| ttx | Constellations could be used as part of the that "documentation" | 09:15 |
| lbragstad | sure | 09:15 |
| ttx | We are not doing a great job at describing those dependencies, so we are not even close to justifying why they are useful | 09:15 |
| * ttx catches up on TC-related ML threads | 09:16 | |
| * ttx refrains from replying to every message | 09:16 | |
| evrardjp | ttx: I agree on the costly bit, and that it should be justified | 09:17 |
| evrardjp | or at least documented clearly | 09:17 |
| lbragstad | i can think of a couple examples where there might not be an "if" for operators, and depending on how developers justify it there might not be another other option for operators | 09:18 |
| ttx | Said another way, projects should not feel forced to depend on other OpenStack projects just for the sake of integration. | 09:18 |
| evrardjp | the explicit is also very interesting -- it would have avoided a few ML conversation about project x or y requiring ceilometer or horizon, if you see what I mean | 09:18 |
| cmurphy | I don't quite recall there being much stress on the costly part, at least the feedback from penick was that as long as it's justified it's not a big deal | 09:18 |
| lbragstad | +1 ^ | 09:18 |
| cmurphy | you deploy another puppet module or ansible role and call it good | 09:18 |
| ttx | cmurphy: maybe s/costly/not free/ | 09:19 |
| evrardjp | cmurphy: agreed | 09:19 |
| lbragstad | if the initial response from operators is "no", take the next step in the conversation and start talking about the justification | 09:19 |
| evrardjp | as long as it's explained and they understand the value :) | 09:19 |
| evrardjp | but it was also said that the more dependencies it has, the harder it is to integrate | 09:20 |
| cmurphy | similar to how the discussion went with ops about making fernet tokens default in keystone, if it's justified and documented it's not a big deal | 09:20 |
| lbragstad | i thought the interesting example jroll brought up was the secret storage use-case with barbican | 09:21 |
| evrardjp | for OSA fernet token as default was a quite smooth change -- I have other examples that are a chain of dependencies that are not so smooth. | 09:21 |
| evrardjp | I think it's good to make a difference between base services and others in that kind of conversation | 09:24 |
| evrardjp | base services are generally accepted as a "we must do it" | 09:24 |
| evrardjp | while the others are little more about picking what you need based on the use case | 09:25 |
| lbragstad | yeah - good point | 09:25 |
| evrardjp | if a castellan compatible base service is required a base services and depended on by all services, barbican becomes a de-facto thing that everyone will deploy -- it just becomes a natural thing | 09:26 |
| evrardjp | and very easy to justify | 09:26 |
| lbragstad | if or when services support putting secrets in barbican | 09:26 |
| evrardjp | I could definitely see a positive security improvement that could be justifiable | 09:27 |
| lbragstad | (e.g. keypairs in nova probably isn't the best example, but certificates in magnum might be better) | 09:27 |
| evrardjp | anyway, explicit is better than implicit, so it should apply to dependencies, right? ;p | 09:29 |
| cmurphy | even "base services" aren't always deployed on 100% of clouds, eg there are openstack deployments with no keystone | 09:30 |
| cmurphy | so it still comes back to your use case | 09:30 |
| lbragstad | ok - that makes me lean even harder on constellations | 09:31 |
| lbragstad | if that's the case, migrating certificates from magnum's home-grown backend to barbican would require supporting both ways for a period of time | 09:32 |
| lbragstad | would that be supported forever to allow use-case driven deployment flexibility? | 09:33 |
| *** dtantsur|afk is now known as dtantsur | 09:36 | |
| cmurphy | i guess that would depend on what the magnum team wants to do and what kind of feedback they get from operators | 09:37 |
| lbragstad | yeah... | 09:38 |
| evrardjp | cmurphy: agreed there | 09:43 |
| evrardjp | base services was just used to prove a point for making it easier to understand for operators, not for saying anything about a forced dependency. | 09:45 |
| lbragstad | i was under the assumption base service meant (at least to some extent) that those things would always be available (to developers and operators), but i wouldn't be surprised if i misinterpreted something | 09:46 |
| * dims peeks | 09:47 | |
| cdent | lbragstad: it certainly sounds that way doesn't it? but I think the targets have moved a bit | 09:47 |
| evrardjp | well etcd is a recent addition, and it's not used in any way in OSA yet, as far as I am aware | 09:47 |
| evrardjp | to be honest, I am a little scared of the addition of any base service -- it means a possible increased complexity | 09:49 |
| evrardjp | but that's not the topic of the discussion | 09:49 |
| evrardjp | long story short: I like the fact we explicitly list the depends on from projects -- it helps deployers and operators | 09:50 |
| evrardjp | whether this goes in the scope of constellations or not -- I would leave that to you :p | 09:50 |
| cdent | cmurphy: I'm definitely going with crash: I've been up for two hours and I'm ready to go back to bed | 09:54 |
| cmurphy | cdent: blegh :( | 09:55 |
| evrardjp | cdent: is that related to tc conversations? ;) | 09:56 |
| evrardjp | lbragstad: I mean 'used to prove a point' in this conversation :) | 09:57 |
| evrardjp | meant* | 09:57 |
| cdent | evrardjp: I just looked at the number of reviews and emails I need to attend to, and then there's this "ptg summary" blog post I'm trying to do...and the bed looks a lot more comfy | 09:57 |
| evrardjp | hahah | 09:57 |
| lbragstad | evrardjp yeah - i guess it depends on where that complexity lives and for how long | 09:58 |
| evrardjp | for the dependency conversation, we also expressed the "hard" and "soft" dependency. I like that we make a difference there: soft paves a way of being standalone while still becoming part of an ecosystem. | 09:59 |
| *** e0ne has joined #openstack-tc | 10:04 | |
| *** e0ne has quit IRC | 11:51 | |
| *** dtantsur is now known as dtantsur|brb | 12:40 | |
| *** mriedem has joined #openstack-tc | 13:07 | |
| *** jamesmcarthur has joined #openstack-tc | 13:28 | |
| *** jamesmcarthur has quit IRC | 13:32 | |
| *** lbragstad has quit IRC | 13:53 | |
| *** lbragstad has joined #openstack-tc | 13:54 | |
| mnaser | o/ | 13:59 |
| dhellmann | o/ | 14:03 |
| dhellmann | lbragstad : thanks for your summary post | 14:03 |
| lbragstad | dhellmann no problem - hopefully folks found it useful | 14:04 |
| lbragstad | i don't think i captured everything, we covered a lot | 14:04 |
| dhellmann | you covered the dependency discussion; I missed that | 14:04 |
| dhellmann | I think I didn't take any notes during that section of the meeting | 14:04 |
| dhellmann | if either of us missed anything I'm sure cdent's summary will cover it | 14:05 |
| ttx | lbragstad: where is your summary post? | 14:06 |
| cdent | dhellmann: thanks for the vote of confidence, but I'm struggling to remember much. Currently editing mine, and it is rather content free. It is more about the feeling of the thing than what we actually did. I think I was ill enough by friday that my brain was very squishy | 14:06 |
| cdent | ttx: https://www.lbragstad.com/blog/openstack-stein-ptg-tc-report | 14:06 |
| ttx | hah! | 14:06 |
| * dhellmann wonders if lbragstad is on planet.openstack.org | 14:06 | |
| dhellmann | cdent : I'm sorry you were (are?) feeling so poorly :-/ | 14:07 |
| *** annabelleB has joined #openstack-tc | 14:07 | |
| cdent | it's inevitable, unfortunately | 14:07 |
| lbragstad | dhellmann i think i signed up to have my posts replicated, but i'm not sure if i did things properly | 14:07 |
| cdent | openstack events are my main engagements with the immune system reinforcement pool | 14:07 |
| lbragstad | i remember we needed to communicate deployment changes for making fernet default blog.openstack.org was suggested | 14:08 |
| dhellmann | lbragstad : yeah, I see your "openstack" feed there | 14:08 |
| ttx | lbragstad: but that post was not (yet?) syndicated apparently | 14:08 |
| * ttx checks feeds | 14:09 | |
| *** dtantsur|brb is now known as dtantsur | 14:09 | |
| ttx | yeah https://www.lbragstad.com/category/openstack/feed/ does not resolve | 14:09 |
| ttx | lbragstad: that is what you have as syndicated feed ^ | 14:11 |
| ttx | It 404s | 14:11 |
| lbragstad | mmm - i bet i tried to set that up before i migrated to a different platform | 14:11 |
| cdent | dhellmann: but yeah, still poorly. great fun being on a plane with a cold | 14:11 |
| zaneb | lbragstad: https://www.lbragstad.com/?category=openstack&format=RSS seems to work | 14:12 |
| * dhellmann hands cdent a hot water bottle and a glass of whiskey | 14:12 | |
| cdent | thanks | 14:12 |
| lbragstad | i suppose i need to update my feed then? | 14:12 |
| lbragstad | to point to the new location | 14:13 |
| ttx | lbragstad: yes. repo is openstack/openstack-planet | 14:13 |
| ttx | (file is planet.ini) | 14:13 |
| lbragstad | hopefully https://review.openstack.org/#/c/603403/ does the trcik | 14:15 |
| ttx | +2a | 14:16 |
| lbragstad | ty | 14:16 |
| *** annabelleB has quit IRC | 14:22 | |
| *** annabelleB has joined #openstack-tc | 14:28 | |
| cdent | dhellmann: here ya go, it's not super fun: https://anticdent.org/openstack-stein-ptg.html | 14:54 |
| dhellmann | cdent : thanks for that. the bit about exposing gaps and removing items from priority lists meshes with what i was trying to have the tc do sunday. | 15:06 |
| dhellmann | I wish I knew the story behind Alex's tweet. I saw something from him on the mailing list that makes more sense now. | 15:06 |
| cdent | yeah, it was an interesting parallel, but was less successful in the nova room than the tc room | 15:07 |
| mnaser | do we have more context on this? | 15:08 |
| mnaser | i don't think interrupting by laughing helps in productivity | 15:08 |
| cdent | dhellmann, mnaser: I think you'll have to find out from alex or someone who was in the room, but it is not at all surprising for the third nova day | 15:10 |
| mnaser | "but it is not at all surprising for the third nova day" paints a negative view that you have which echos to the rest of the tc, i'd rather us be more subjective in saying "foo did bar" rather than "#justnovathings" | 15:11 |
| cdent | the "miscellaneous" category is always dangeours territory | 15:11 |
| *** jamesmcarthur has joined #openstack-tc | 15:12 | |
| cdent | mnaser: I've given up trying to paint a picture of nova for the sake of eveyone else. I've got my opinions and that's all I'm able to share | 15:12 |
| mnaser | i think that's a bit problematic, given that your position on the tc puts you in a figure of authority, saying "i think nova sucks" = "tc thinks nova sucks" for the majority of the community imho | 15:13 |
| cdent | I agree it is problematic that nova sucks for some people. | 15:14 |
| smcginnis | I can't get behind any idea that the TC members must forfiet all independent thought to be on the TC. | 15:14 |
| cdent | If I'm supposed to keep my mouth shut about the problems I experience to preserve some kind of false peace, then I'd rather not be on the TC | 15:14 |
| mnaser | i don't think that's what i'm trying to say | 15:15 |
| cdent | I was elected to the tc in large part because I'm willing to speak out about things like nova sometimes sucking | 15:15 |
| mnaser | what i'm saying if there is a *specific* issue we need to address, let's talk about it. if alex was interrupted by laughtor in the room, then let's talk about that issue. | 15:15 |
| mnaser | it's not because ~nova team sucks therefore anything happening in that room on friday sucks~ | 15:15 |
| cdent | all I'm doing by saying that is that it is part of a trend | 15:16 |
| cdent | yes we should address the specific instance | 15:16 |
| cdent | but we should also address that it is a part of a trend | 15:16 |
| cdent | the trend _matters_ | 15:16 |
| mnaser | imho people/group issues can't be trended on a chart and fixed like a technical issue | 15:16 |
| cdent | and fixing the symptoms does not address the trend | 15:17 |
| mnaser | there are many external factors affecting that | 15:17 |
| mnaser | and resolving each issue in isolation makes it much better than alienating a group of people as 'trendingly bad' | 15:17 |
| cdent | i'm not asking that things be fixed like a technical issue. In fact I'm explicitly asking that we stop trying to address people problems as if they are technical issues. | 15:17 |
| mnaser | if anything that would probably add a "hey, screw ya, if you think i'm always a problem, i'll just ignore you" | 15:17 |
| mnaser | i don't think it's fun being on a team that is labeled as a "problem team" all teh time | 15:18 |
| cdent | I'm trying _really_ hard to make it clear that this is a system-oriented issue wherein the forces that press on nova result in behaviors that are less than ideal. It is _not_ because of the individuals. | 15:18 |
| mnaser | which arguably delivers the most one of the biggest openstack deliverables, consistently and succesfully | 15:18 |
| mnaser | but instead "they have a trend of sucking, so they suck, nothing new" | 15:19 |
| cdent | It does, sure, but at what cost? | 15:19 |
| mnaser | right but you're speaking on *behalf* of the nova team now | 15:19 |
| mnaser | that these system-oriented issues are a thing | 15:19 |
| cdent | how am I speaking on behalf of the nova team? | 15:19 |
| cdent | I have never felt like a part of the nova team | 15:20 |
| dtroyer | FWIW, I was in the Nova room during the discussion of Alex's item, it included a number of comments about the state of Intel's 3rd party CI (not great) and more related to that tweet some feelings toward what some see as a marketing checkbox item. | 15:20 |
| cdent | which is part of the problem | 15:20 |
| mnaser | "this is a system-oriented issue wherein the forces that press on nova result in behaviors that are less than ideal" | 15:20 |
| mnaser | the only person who determines that is someone who's on that team | 15:20 |
| smcginnis | So now we can't even make observations? | 15:20 |
| mnaser | i'm all for observations. as long as they are objective | 15:21 |
| smcginnis | I'm not on the nova team but I can agree that there are system oriented issues. | 15:21 |
| cdent | mnaser: I'm completely at a loss for what you are trying to say/prove | 15:21 |
| cdent | there is no such thing as an objective observation by a human | 15:21 |
| mnaser | not "it was a friday at the nova ptg room so it wouldn't surprise me that people laughed out loud at something" | 15:21 |
| mnaser | in other words: "the nova team is a bunch of insensitive people and it doesn't surprise me they'd do this on the last day of the ptg" | 15:22 |
| cdent | there have been four friday's at the ptg in the nova room. in each one people who are providing miscellaneous topics have been responded to with lack of attention and what I would call disrespect | 15:22 |
| mnaser | that's how most people parse that | 15:22 |
| mnaser | so let's bring *those* cases up, objectively, and decide why we think that's the case | 15:22 |
| cdent | omg | 15:22 |
| cdent | how long do you think I've been trying to address these issue mnaser ? | 15:22 |
| cdent | how many times do you think I've brought up individual issues? | 15:23 |
| mnaser | i don't see anything besides placement | 15:23 |
| mnaser | in the recent history | 15:23 |
| cdent | wow | 15:23 |
| mnaser | i don't know where they've been brought up | 15:23 |
| mnaser | but i certainly don't recall any discussions besides nova on my time at the tc | 15:23 |
| cdent | how about just take me on faith? | 15:23 |
| cdent | I'm sick to death of being treated as some kind of whiny troublemaker when I'm one of the few people who have been willing to stick around and speak up about this stuff | 15:24 |
| cdent | I don't have to care about this stuff | 15:24 |
| mnaser | i welcome the fact that you push the envelope and speak about things that we don't want to talk about | 15:24 |
| cdent | and the response I get from people makes it seem like I'm wasting my time | 15:24 |
| mnaser | but i think we need to talk more specifically about specific topics (i.e. placement, alex issue), rather than promote 'the nova team clearly sucks and will never improve' type of messaging | 15:25 |
| cdent | I alread said: If you want more info from alex's situation: talk to him | 15:25 |
| cdent | but don't expect me not amplify his signal | 15:25 |
| mnaser | i don't expect you, but you're also amplifying the signal that "it's friday and it's nova so nothing new here about them being rude" | 15:25 |
| cdent | a) I haven't said things will never improve, but b) do you understand why I might, over sufficient time, start to feel that way | 15:26 |
| cdent | mnaser: I said that in here to provide more context to you and dhellmann. That lack of surprise is an important data point, I think. | 15:27 |
| mnaser | i can get why. but we need to have the conversations with everyone present | 15:27 |
| mnaser | and talk about the 'issues', rather than the generic 'nova sucks' messaging | 15:27 |
| cdent | I think you jumped to conclusions in this case, based on conversations we've had elsewhere. | 15:28 |
| cdent | I'd like to be able to provide context to issues that are raised without you jumping to those conclusions. | 15:29 |
| cdent | In my blog post, I just point out the issue and try not to make conclusions about why. | 15:29 |
| cdent | Here we started a conversation and you pretty much instantly tried to shut me down | 15:30 |
| cdent | That's not okay. | 15:30 |
| mnaser | i'm not shutting anyone down. i said i'm all for bringing up issues, i'm not for labeling projects as "bad actors" in a geeneric fashion | 15:30 |
| cdent | And I don't think I did that here. | 15:31 |
| smcginnis | cdent is not the one I saw making that conclusion, FWIW. | 15:31 |
| mriedem | oh i guess i should have been paying attention in the tc channel once again | 15:31 |
| mnaser | so your messaging doesn't really paint the nova team badly in any way by saying "it's friday and someone laughing about something is totally not normal" | 15:32 |
| * smcginnis steps away for coffee and something more productive | 15:32 | |
| * persia notes that meetings with everyone present are a lovely idea, but conversations in stadiums don't tend to be inclusive | 15:32 | |
| mnaser | persia: another reason for meetings, yay! :> | 15:33 |
| mnaser | (i'm supportive of them) | 15:33 |
| mriedem | i would just hope that when it's time for shitting on nova without any nova people around, someone would notify | 15:33 |
| mriedem | "hey it's time to get defensive" | 15:33 |
| cdent | jesus, nobody shat on nova | 15:34 |
| mriedem | "but it is not at all surprising for the third nova day" | 15:35 |
| cdent | mriedem: this started because of me including alex's tweet in https://anticdent.org/openstack-stein-ptg.html | 15:36 |
| cdent | and I observed that, in my experience, friday's in the nova room often involve some disrespect | 15:36 |
| cdent | is that shitting on nova? | 15:36 |
| mriedem | i don't remember people laughing at alex | 15:36 |
| mnaser | since i got on the tc, there's a systematic messaging against the nova team, maybe not explicitly but implicitly all the time | 15:36 |
| cdent | I would have thought it was shitting on the third day in nova | 15:36 |
| cdent | my perspective is that the third day is intense and difficult because there's too much pressure | 15:37 |
| mriedem | i thought friday was pretty laid back myself and ended early | 15:38 |
| zaneb | mriedem: I seem to recall you making a speech in which people in the Nova room were said to be "strangling each other" by the third day | 15:38 |
| mriedem | oh yes i was being very literal | 15:38 |
| mriedem | maybe this doesn't happen to other projects, | 15:39 |
| mriedem | by friday is mostly our smorgasbord day, | 15:39 |
| mriedem | where we have a giant etherpad full of random requirements, | 15:39 |
| mriedem | and by the end of friday, it's exhausting | 15:39 |
| *** aprice has joined #openstack-tc | 15:39 | |
| mriedem | and i do tend to speak directly / frankly to avoid hemming and hawing on a fence about something for 2 hours | 15:39 |
| mriedem | that can hurt peoples feelings, | 15:39 |
| zaneb | mriedem: my point is that there doesn't seem to be a great deal of difference between how you and cdent describe day 3 of a Nova PTG | 15:39 |
| mriedem | but it also hurts peoples feelings when we don't give an answer and then just ignore their thing for a year | 15:40 |
| mriedem | cdent isn't providing context for the people that weren't in the room | 15:40 |
| mriedem | is the difference | 15:40 |
| * cdent blinks | 15:41 | |
| cdent | I admit that in my quoting | 15:41 |
| mriedem | dhellmann said, "I wish I knew the story behind Alex's tweet." | 15:41 |
| mriedem | i wish more tc people would ask that | 15:41 |
| cdent | me too | 15:41 |
| mriedem | than just assuming "trollers gonna troll when it comes to nova" | 15:41 |
| * cdent sighs | 15:42 | |
| mnaser | mriedem: that was my point there. | 15:42 |
| mnaser | that's all it was | 15:42 |
| zaneb | I did some discreet research on my own (i.e. digging through IRC logs) but wasn't able to come up with anything | 15:42 |
| ttx | ftr "I wish I knew the story behind Alex's tweet" too | 15:42 |
| ttx | and will be reaching out to him to get a less cryptic idea of what happened | 15:43 |
| cdent | We spent much of last week talking about visibility being a useful way for the right things to get attention. I saw that tweet and resonated strongly with me, so I quoted it. Does there have to be more to it than that? The rest of the blog post I explain _my_ feeling about why the week is hard. I don't know alex's reasons and I said so. | 15:43 |
| ttx | (from his persepctive) | 15:43 |
| mnaser | i was going to try and reach out to folks involved as well as alex and see what the issue is too | 15:44 |
| cdent | I'm not attempting to trash nova, or openstack, or the tc, or anything else. I'm trying to give us a chance to think about things from many perspectives. Loads of people come out of the ptg fired up and positive. I did not. So did some other people. That's a useful data point, I think. | 15:44 |
| cdent | So to that extent I'm glad I quoted the tweet so now mnaser, ttx, and dhellmann might have some energy to investigate | 15:45 |
| cdent | thtat's good | 15:45 |
| mnaser | cdent: but that's not the issue, all of that is fine, it's the point mriedem just mentioned | 15:45 |
| cdent | tell me again what _I_ did that is the issue? | 15:45 |
| mnaser | i liked the blog post / article and i went through it all, it's very informative. you even praised members of the nova team in there at some point for their work regarding the whole placement stuff | 15:45 |
| mnaser | mriedem: last few messages summarized it for me | 15:46 |
| cdent | I'm asking you to rephrase what mriedem said for my sake | 15:46 |
| mriedem | tbc, what dtroyer said before: "(10:20:22 AM) dtroyer: FWIW, I was in the Nova room during the discussion of Alex's item, it included a number of comments about the state of Intel's 3rd party CI (not great) and more related to that tweet some feelings toward what some see as a marketing checkbox item." | 15:47 |
| mriedem | i'm the one that called out the marketing checkbox | 15:47 |
| mnaser | the comment of "but it is not at all surprising for the third nova day" gives out a message that "trollers gonna troll when it comes to nova". rather than us trying to seek out what happened | 15:48 |
| mnaser | that's all | 15:48 |
| mriedem | because intel comes with a laundry list of complex requirements to every ptg saying "our hardware can do this so nova should enable it" | 15:48 |
| mriedem | and we try to be very nice and smile and just hope it goes away because no one outside of intel wants to work on that stuff | 15:48 |
| mriedem | we = me | 15:48 |
| mnaser | mriedem: i assume the lack of a stable ci and intel contributors means that the list isn't as productive? | 15:48 |
| mriedem | those are table stakes | 15:49 |
| mnaser | i'm just guessing | 15:49 |
| mriedem | if you want to push something that no one else cares about and operators/users aren't asking for, those are table stakes for me to maintain your unicorn | 15:49 |
| mnaser | are these proposals that come with contributors or just proposals? | 15:49 |
| mriedem | i'm sure there would be a short term contributor to get the thing in | 15:50 |
| mriedem | it's especially difficult to see this all going through alex who is a nova core | 15:50 |
| mnaser | how so? | 15:51 |
| mriedem | credibility? | 15:51 |
| mriedem | this is why i said to him, directly, in the room, "i don't have a problem with *you*, i have a problem with that your company is pushing on you" | 15:51 |
| mriedem | s/that/what/ | 15:51 |
| mnaser | i think that might have been the hardest thing on him | 15:52 |
| mnaser | on one side his employer having certain requirements, but on the other the downstream teams refusing them (i guess?) | 15:52 |
| mriedem | i go through the exact same stuff | 15:52 |
| mriedem | i can't imagine i'm the only one that has downstream product people wanting me to push things upstream | 15:53 |
| mnaser | i mean don't most companies run intel hardware so isn't there benefits in having whatever those features are (i am not familiar with what they are tbh) | 15:53 |
| mriedem | the thing being pushed last week is not even GA yet | 15:53 |
| mnaser | well maybe we should more clearly communicate that we can't add features for things that are not GA? | 15:54 |
| jroll | I often see vendors trying to push features onto upstream services that aren't really out in the wild yet | 15:54 |
| mriedem | cdent: even in your blog post you say, "The trick now is to follow through and focus on those things while avoiding adding yet more to the list." - that is precisely why i get terse with the laundry list | 15:54 |
| jroll | I assume they're hoping that the inclusion of the feature upstream can generate sales | 15:54 |
| mnaser | maybe this is something the tc should talk about saying that contributions involving other tools/software/etc must be GA? | 15:55 |
| mriedem | we (nova) will definitely not get done the big things we need to get done if we're constantly distracted with the unicorn features | 15:55 |
| mnaser | that way alex would be able to communicate more clearly with his upstream team that "this isn't the right time" | 15:55 |
| mriedem | jroll: "I assume they're hoping that the inclusion of the feature upstream can generate sales" i assume the same | 15:56 |
| cdent | mriedem: yeah, and as we talked sometime during the week, I've never disagreed with saying no, but the mode in which is happens is problematic. I'm not sure how to fix it, but we can do better somehow. We don't want people going off feeling like they never want to come back. | 15:56 |
| mnaser | i would much rather have them know "this isn't a good time" | 15:56 |
| jroll | mnaser: it's less about it being the right time, and more about trying to prioritize things that users/ops are asking for, rather than things the vendor hopes they'll ask for one day | 15:56 |
| mnaser | jroll: well, if they show up with contributors to add it and it's relatively maintainable in it's own right (i'm just guessing here, like, imagine some sort of driver or localized functionality) then it's not too much of an issue | 15:57 |
| ttx | maybe a clearer categorization of the request (strategic vs. tactical) would help | 15:57 |
| mnaser | but if it's to integrate something that doesn't even exist yet in GA.. *shrug* | 15:57 |
| mnaser | (fyi, these are the conversations that we should be having to help manage those expectations, this is productive and thanks for adding insight mriedem) | 15:58 |
| ttx | we can't help with your tactical request because we are swamped with higher-prio strategic things that benefit everyone | 15:58 |
| jroll | mnaser: the craziness that vendors are doing these days tends to be a bit more further reaching... the latest flavor with ironic and neutron is NICs that run OVS O_O | 15:58 |
| mnaser | jroll: that sounds pretty badass | 15:58 |
| jroll | it's insane IMO | 15:59 |
| mnaser | makes baremetal so much more uasble | 15:59 |
| mnaser | now i want to know whos that vendor :p | 15:59 |
| jroll | intel | 15:59 |
| zaneb | mriedem: suggestion: it sounds like alex may feel caught in the middle because the rejection happens in-person at the PTG, and he is the one who has to go back to his company and explain that he 'failed'? Maybe those situations can be helped by having somebody else write up the reasons for rejecting something on the ML that he can point to instead of being stuck in the middle? (I realise this is more work for | 15:59 |
| zaneb | someone ): | 15:59 |
| mnaser | switches can go back to being stoopid | 15:59 |
| mnaser | but again, until those nics out in the wild, we shouldn't be adding code to neutron or ironic to support them... | 15:59 |
| mnaser | not sure how you feel about that statement jroll ^ | 15:59 |
| jroll | mnaser: the only reason I can get behind moving on it is that neutron is doing architecture to support it, in a way that makes it better for other use cases | 15:59 |
| mriedem | zaneb: my comments on the thing in question are in the etherpad | 16:00 |
| mriedem | alex can clearly point to that | 16:00 |
| jroll | mnaser: sounds great until there's a CVE in OVS and you have to go update a fleet worth of firmware | 16:00 |
| jroll | anyway | 16:00 |
| *** jamesmcarthur has quit IRC | 16:00 | |
| mnaser | jroll: yeah, good and bad out of it | 16:00 |
| mriedem | L957 https://etherpad.openstack.org/p/nova-ptg-stein | 16:00 |
| zaneb | mriedem: ok, fair enough. I saw mentioned in the Twitter thing that he was e.g. asking questions he knew the answer to on IRC in order to explain to people internally | 16:01 |
| jroll | I agree that things should be GA or in the wild before we support them, but I fear if we write down that rule, people will say "look it's GA, merge now pls" and then we go back to the priority conversation, which is the main issue | 16:01 |
| mnaser | jroll: well, it at least filters *some* of the issues though | 16:01 |
| ttx | saying "your request belongs in category B"Â might feel less personal than " I have a problem with your company | 16:01 |
| ttx | " | 16:02 |
| mnaser | ttx: those are my thoughts | 16:02 |
| jroll | mnaser: IMO it just puts them off | 16:02 |
| mnaser | and it's easier to go back to your employer and say "hi, we cant do this because policy X says so" | 16:02 |
| ttx | mnaser: right | 16:02 |
| mnaser | rather than "they disagreed / i couldn't convince them" | 16:02 |
| mriedem | so we've once again come around to solving these people things with policies | 16:02 |
| mriedem | and rules | 16:02 |
| mriedem | technical things to hide behind | 16:02 |
| cdent | "technical things to hide behind" is a concern I share | 16:03 |
| mriedem | rule IV-X-4 says we can reject your snowflake on these grounds | 16:03 |
| ttx | mriedem: I would not use "policy" to say no. But having general agreement that tactical requests have less priority than strategic requests sounds like a good idea | 16:03 |
| mriedem | ttx: and then we'll continue to say that, | 16:04 |
| jroll | this is where "I have a problem with your company" comes from: https://governance.openstack.org/tc/reference/principles.html#openstack-first-project-team-second-company-third | 16:04 |
| *** jamesmcarthur has joined #openstack-tc | 16:04 | |
| jroll | > in case of conflicts of interest, they should be ready to put those needs aside and make the best call for OpenStack as a whole | 16:04 |
| mriedem | to the point that their request is just abandoned | 16:04 |
| mriedem | that's the long-term goal in saying that | 16:04 |
| mriedem | and i'm saying, just be honest up front | 16:04 |
| mriedem | "no one wants this" | 16:04 |
| jroll | sounds like intel is not accepting that for alex, and being called out, as they should | 16:04 |
| * mnaser is going to run OSA meeting for an hour bbl | 16:04 | |
| ttx | well "no one but you" :) | 16:04 |
| mriedem | "no one but your sales team" | 16:04 |
| mriedem | i believe that's what i said in the ptg room last week | 16:05 |
| mriedem | i guess i'm a monster | 16:05 |
| zaneb | jroll: the flip side of that is that we should assume people are following that in good faith - i.e. the people like Alex who are sent out into the community are on *our* side | 16:05 |
| ttx | mriedem: I think we are saying the same thing. No one except your sales team wants this, so that makes it a low priority item, and we have plenty of high priority work to focus on" | 16:06 |
| mriedem | so it's cool as long as i don't say it | 16:06 |
| mriedem | noted | 16:06 |
| mriedem | i'll reach out to alex over email, i tried getting him on irc earlier today but he was already gone (or hates me still) | 16:07 |
| mriedem | since i'm not on the tweets | 16:07 |
| jroll | zaneb: sure, I don't tend to assume good faith from corporate internals though :) | 16:07 |
| ttx | I'm not having an opinion on that (cool or not cool) until I have more context anyway | 16:08 |
| ttx | but explaining it in the context or priorities (less people want it = less priority) sounds like a good way to communicate why 'no' | 16:08 |
| ttx | s/or/of | 16:09 |
| jroll | I really don't think any of this is about nova blocking a feature or explaining the block "wrong" | 16:09 |
| jroll | it appears the laughter and the interruption was the issue | 16:09 |
| jroll | which is not surprising in a PTG setting, lots of jokes and interruptions there in general, intentional or not | 16:10 |
| mnaser | jroll: there was a few follow up tweets tho | 16:10 |
| mnaser | not sure if you saw them too | 16:10 |
| jroll | mnaser: yep | 16:11 |
| jroll | I interpreted the whole thing as "I'm just trying to do my job, it's a really hard one, and y'all are going to interrupt me and make jokes?" | 16:12 |
| cdent | jroll: yeah | 16:12 |
| mnaser | anyhow i'm glad mriedem is taking initiative in reaching out to alex personally | 16:12 |
| jroll | I'm not saying the laughter and interruption is okay, I'm just saying that it isn't surprising (and this isn't limited to the nova team, despite how it was said/perceived an hour ago) | 16:13 |
| jroll | now I think back to the "ugh"s and such I've muttered in PTG sessions for features I don't think should be in ironic or don't care about or whatever :/ | 16:13 |
| jroll | I'm sure everyone here has been on at least one side of this | 16:14 |
| mriedem | i pushed db2 support into nova for 2 years | 16:14 |
| mriedem | you can probably imagine there were some jokes made along the way | 16:14 |
| evrardjp | mriedem: All becomes clear now. | 16:15 |
| evrardjp | (adding jokes on top of it to prove a point) | 16:15 |
| mriedem | was that a joke? | 16:15 |
| * mriedem opens twitter account | 16:15 | |
| cdent | normal != okay | 16:16 |
| cdent | normal != not worth improving | 16:16 |
| cdent | but at the same time just because stuff can be improved doesn't mean it is horrible nasty full of blame a disaster and everyone needs to be censured | 16:17 |
| cdent | it just means it can be improved | 16:17 |
| jroll | ++ | 16:18 |
| evrardjp | okay to come back to the point, thanks mriedem indeed to talk personally on that | 16:18 |
| jroll | fwiw, I didn't read the original statement as an attack on nova, only an observation that I have also observed in nova *and* on other teams | 16:19 |
| ttx | jroll: yes the root frustration is being sent to convince a bunch of people to do something they are not really interested in, so maybe some people high up need a bit more education | 16:20 |
| zaneb | for the record I am quite sure that I have been a jerk to people when rejecting their pet feature | 16:21 |
| evrardjp | ttx: the question I wonder is that -- how often has that happened? Only one is too much for diluting the community | 16:21 |
| ttx | which is what I alluded to when I said categorization... I was not talking about hiding behind a process to say no. but about justifying no by pointing back to our principles (project first) | 16:22 |
| evrardjp | oh I understand what you mean | 16:22 |
| ttx | evrardjp: I think some orgs thinking they can use "the community" to do their bidding is happening daily | 16:22 |
| smcginnis | Open source is just about getting other people to do your work for you, right? :) | 16:23 |
| ttx | that puts people in awkward positions, and sometimes it explodes | 16:23 |
| evrardjp | that's true, but if there is a risk of lack of company diversity preventing things to change, we also would need to be aware of it | 16:23 |
| ttx | If we could build more of the right expectations in those orgs management chains, maybe we'll have less of that awkward | 16:24 |
| jroll | ttx: we all (most?) deal with corporate overlords asking us to do things that may conflict with the community. it's part of the job. I think the problem here is that people feel like they're being ridiculed for doing said job. | 16:24 |
| evrardjp | ttx: wow that's a hard goal | 16:24 |
| ttx | We solved all the easy problems already | 16:24 |
| evrardjp | it's an everyday job :) | 16:24 |
| ttx | :P | 16:24 |
| evrardjp | jroll: yeah that's definitely something that should not happen and can be improved -- I am not aware of the root cause so I will stop speaking right there | 16:25 |
| jroll | like yes, I applaud your efforts to fix the insides of the corporations, but I also think that's a people problem and there's no policy or technical solution that can do such a thing | 16:25 |
| evrardjp | agreed | 16:26 |
| jroll | evrardjp: I think it's that we're all humans (or mostly humans) with varying length of fuses :) | 16:26 |
| ttx | jroll: people's problems can be fixed by education | 16:26 |
| evrardjp | but on that point we already have an action item with mriedem | 16:26 |
| ttx | "fixing insides of corporations" is an education thing | 16:27 |
| evrardjp | so that's why I moved on -- and hope to have a positive outcome | 16:27 |
| *** dtantsur is now known as dtantsur|afk | 16:28 | |
| *** jpich has quit IRC | 16:33 | |
| *** annabelleB has quit IRC | 16:38 | |
| *** annabelleB has joined #openstack-tc | 16:40 | |
| persia | Maybe it would help to promote work-exchange discussions in our contributors? If someone has a pet feature, and they need help from others to get it in, to spend time working on others' pet features in exchange for the help? | 16:41 |
| cdent | that's already de rigueur persia | 16:43 |
| persia | yes, but does everyone know that? | 16:43 |
| cdent | good question | 16:43 |
| persia | I saw a number of folk last week who were trying to communicate things that needed doing to other folk. I saw fewer folk offering to do things in exchange for help. | 16:44 |
| persia | (by and large, the latter received prompt attention) | 16:44 |
| cdent | I reckon that goes back to: lots of people over-extended | 16:44 |
| persia | Oh, probably :) | 16:45 |
| jroll | we've often explicitly told vendors in ironic-land that if they want their feature to be prioritized faster, they can help us review code so that we get the higher priorities done more quickly | 16:45 |
| jroll | it took a long time of repeating that for vendor folks to understand it | 16:45 |
| persia | jroll: It is unfortunate to say that the ironic prioritisation meetings have long included much more explicit work-exchange discussion than is common for other projects. Part of me chalks that up to how Ironic is, at base, dependent on specific vendor support, but there may also be some different techniques that have been used by Ironic PTLs over the years. | 16:47 |
| jroll | persia: I would agree with that. probably some of both :) | 16:47 |
| persia | I also remember a sequence of Cinder meetings in which many folk suddenly "got it", after which things seemed less painful. | 16:47 |
| persia | (for some of the same reasons) | 16:48 |
| jroll | nod | 16:48 |
| persia | Unfortunately, I've only attended that class of discussion for maybe 10% of our projects, so I don't know how much translates widely. | 16:48 |
| knikolla | o/ | 16:49 |
| mnaser | is there a list of the newton ptls | 17:13 |
| mnaser | looks like the site goes as back to ocata | 17:13 |
| ianychoi | mnaser, previous lists are on Wiki - I think https://wiki.openstack.org/wiki/PTL_Elections_March_2016 has Newton PTL lists | 17:15 |
| ianychoi | ( http://git.openstack.org/cgit/openstack/governance/refs/?id=march-2016-elections would be better ) | 17:16 |
| persia | Just as a note: those tags are just *before* elections, rather than just after, so one has calculate the dates differently. | 17:18 |
| mnaser | yeah i saw the tags but they're pre-election | 17:19 |
| mnaser | thanks ianychoi ! | 17:19 |
| ianychoi | mnaser, np :) persia oh correct thanks :) | 17:20 |
| persia | mnaser: Maybe http://git.openstack.org/cgit/openstack/governance/tree/reference/projects.yaml?h=sept-2016-elections ? | 17:21 |
| persia | Note that if you want to carefully track who was PTL in cases where folk changed throughout the cycle, you'd need to use more advanced features of git than cgit supports. | 17:22 |
| persia | http://git.openstack.org/cgit/openstack/governance/commit/reference/projects.yaml?h=sept-2016-elections&id=c1923a32923763568b7b9f704f80c49215cc1795 is most of it, although there are some other updates in the log. | 17:23 |
| dhellmann | lbragstad : regarding the tool to use to write up the rbac work; what state is that in now? is there a lot of detail or is it still early planning stages? | 17:40 |
| lbragstad | dhellmann that's a good question... prior to the PTG last year i was using the Ongoing directory of our specifications repository | 17:41 |
| *** jamesmcarthur has quit IRC | 17:41 | |
| dhellmann | ok, specs seem like one of several reasonable options | 17:41 |
| lbragstad | but - other than that the information is scattered across existing specifications, general documentation, and my brain | 17:41 |
| dhellmann | so it would probably be a good idea to pull together a "home page" of some sort for it | 17:42 |
| dhellmann | maybe an etherpad is a good place for that | 17:42 |
| dhellmann | or even a wiki, I guess | 17:42 |
| dhellmann | that can be used to pull together links to everything that exists now | 17:42 |
| lbragstad | for context - this was the first document i started http://specs.openstack.org/openstack/keystone-specs/specs/keystone/ongoing/policy-goals.html | 17:42 |
| lbragstad | and then i tried to make one specific to how that impacts security http://specs.openstack.org/openstack/keystone-specs/specs/keystone/ongoing/policy-security-roadmap.html | 17:43 |
| lbragstad | we also have https://trello.com/b/bpWycnwa/policy-roadmap | 17:44 |
| dhellmann | then if there are multiple phases they can be put in order with a brief description | 17:44 |
| lbragstad | ok | 17:44 |
| dhellmann | I think I would try to limit this to the normal tools | 17:44 |
| dhellmann | maybe that trello board could be imported into storyboard | 17:44 |
| lbragstad | yeah - we really only used it to link to existing specs in other roadmaps | 17:45 |
| dhellmann | yeah, ok | 17:45 |
| lbragstad | but there shouldn't be anything specific to trello for those work items (?) | 17:45 |
| dhellmann | cool | 17:46 |
| dhellmann | and once there's a home page-ish thing for the work, then I would add some info to the community-goals etherpad | 17:46 |
| lbragstad | afaik, we don't really use a wiki any more | 17:47 |
| lbragstad | would our developer documentation be too rigid for something like this? | 17:47 |
| dhellmann | docs or specs repos would be ok, but there's the review process which makes minor updates take longer. it's up to you, though, so use whatever feels like it's meeting the needs you have | 17:48 |
| lbragstad | ok - that makes sense | 17:48 |
| *** diablo_rojo has joined #openstack-tc | 17:57 | |
| dhellmann | tc-members: the board is meeting in ~2 hours. The agenda includes a presentation about the strategic project governance work the foundation has been doing. https://wiki.openstack.org/wiki/Governance/Foundation/18Sep2018BoardMeeting | 17:58 |
| cdent | I want to go, but I'm sure I'm going to be awake | 17:59 |
| zaneb | there goes the afternoon | 17:59 |
| cdent | not sure, of course | 18:05 |
| cdent | I wonder if anyone has coined the phrase "taking a Linus" yet? | 18:05 |
| dhellmann | He's getting an awful lot of credit for coming around on an issue folks have been complaining about for years. I mean, I'm happy he's dealing with it, but still. | 18:07 |
| cdent | indeed | 18:07 |
| notmyname | dhellmann: oh, is that a real board meeting? the email message subject said it was "Training session reminder", and there wasn't an agenda linked | 18:09 |
| dhellmann | notmyname : yeah, they use a webex training tool to control who can talk and present | 18:09 |
| dhellmann | and I guess generating the calendar thing through that makes it come out like it's a training session | 18:10 |
| dims | thanks for the reminder dhellmann | 18:12 |
| * zaneb always wondered why they had so many training sessions | 18:23 | |
| *** ricolin has joined #openstack-tc | 18:46 | |
| mrhillsman | is there a simple way to get - https://docs.openstack.org/nova/latest/reference/api-microversion-history.html - in like a table/json/yaml for each project | 18:47 |
| mrhillsman | version mapping per release, not sure what you would call it | 18:47 |
| smcginnis | mrhillsman: A way to programmatically load that information? | 18:50 |
| mrhillsman | that would work i think | 18:50 |
| mrhillsman | or if there is/was a page with release to project to apiversion map/table | 18:50 |
| smcginnis | Nothing that I am aware of right now. | 18:52 |
| smcginnis | At least in Cinder, the only semi-structured way we have that info is in the api-microversion-history RST file. | 18:53 |
| mrhillsman | i did not see one for cinder | 18:53 |
| smcginnis | That could be parsed, making some assumptions about consistency of format across projects that have similar pages. Not the best solution. | 18:53 |
| mrhillsman | found it | 18:53 |
| smcginnis | https://docs.openstack.org/cinder/latest/contributor/api_microversion_history.html | 18:54 |
| *** annabelleB has quit IRC | 18:56 | |
| *** annabelleB has joined #openstack-tc | 19:00 | |
| fungi | not really around today but trying to catch up on hundreds of lines of scrollback... the historical intel/nova example which sticks with me is the trusted compute scheduler filter, which was completed just enough for the vendor to produce some glossies and a whitepaper describing how their hsm was "supported" in openstack as a selling point for the hardware, even though it was at best a proof-of-concept | 19:01 |
| fungi | implementation they abandoned and left as an attractive nuisance until the nova team (slowly) deprecated it. i can see how things like that leave a negative impression over time. i'm guessing nova regulars have plenty more examples of the same | 19:01 |
| smcginnis | How's the homestead fungi? | 19:02 |
| fungi | intact! | 19:02 |
| fungi | (and not appreciably wet) | 19:03 |
| fungi | the storm mostly swerved to miss us | 19:03 |
| fungi | er, i meant s/hsm/tpm/ above | 19:03 |
| fungi | tla madness ;) | 19:04 |
| smcginnis | fungi: Cool, that's good news. Glad to hear it was better than expected! | 19:05 |
| dims | fungi : excellent news! (homestead) | 19:23 |
| mriedem | mrhillsman: there is a project / release / api versions mapping somewhere | 19:42 |
| mriedem | i remember updating it after each release | 19:42 |
| mriedem | e.g. https://review.openstack.org/#/q/status:merged+project:openstack/project-navigator-data+branch:master+topic:nova-pike-version | 19:43 |
| mriedem | apparently no other projects cared about that since ocata | 19:43 |
| mriedem | or just, no one cares, period. | 19:44 |
| mrhillsman | :( | 19:47 |
| smcginnis | Apparently - https://review.openstack.org/#/q/project:openstack/project-navigator-data+status:open | 19:47 |
| smcginnis | The need to remember to go elsewhere to update that information probably is not the right answer though. | 19:48 |
| smcginnis | Especially as PTLs transition, if it's this separate thing that someone needs to remember to do, it probably won't happen. | 19:48 |
| smcginnis | It might be better to keep a yaml file in each repo at an expected location and have something that is smart enough to find and extract that info. | 19:49 |
| smcginnis | Or given a unique file name, automation could even be used with codesearch.o.o to just find all of them and pull it together in a consumable format. | 19:49 |
| dhellmann | that makes validation a little more complicated; I wonder if we could start with just a reminder to update the file? | 19:49 |
| smcginnis | First question though is - do we still need to update that file? Does anyone care? | 19:50 |
| dhellmann | yeah, that's a good place to start | 19:50 |
| ttx | smcginnis: the answer would be...no | 19:50 |
| ttx | That's a thing that always looked a bit too low-level for the project nav | 19:50 |
| smcginnis | Given no one has been jumping up and down about how out of date it is, I would agree. | 19:51 |
| ttx | I've been campaigning to stop displaying it | 19:51 |
| smcginnis | ++ | 19:51 |
| ttx | If that's useful data, it should be published to doc | 19:51 |
| smcginnis | I think I'd rather see something like the history page pulled into the api-ref documentation. | 19:51 |
| dhellmann | it feels like the sort of data we ought to be able to generate, too | 19:51 |
| ttx | oh yes, and actually it was removed in last project-navigator version | 19:51 |
| ttx | so my campaign was successful | 19:52 |
| smcginnis | Oh? Guess I can abandon my patch. | 19:52 |
| dhellmann | progress! | 19:53 |
| mrhillsman | something machine readable would be great wherever it is | 19:53 |
| ttx | I'm not saying we should stop keeping track of it, just that the project-navigator no longer needs it | 19:54 |
| ttx | the way it was displaying it was more confusing than useful | 19:55 |
| ttx | people thought that was software version and not API version | 19:55 |
| mrhillsman | looking at the pike nova.json it would be nice to have that per project | 19:57 |
| * dhellmann moves over to #openstack-board for the meeting | 19:57 | |
| ttx | smcginnis: maybe start a thread about it ? Saying the data was refreshed only by one team, and the project navigator no longer displays it, so should we care, and if yes where should we display it | 19:59 |
| ttx | IMHO that data is prone to go stale... If we keep it it should be somehow autogenerated | 20:00 |
| *** gouthamr has quit IRC | 20:00 | |
| smcginnis | This at least triggered me to do https://review.openstack.org/603486 | 20:01 |
| smcginnis | ttx: I'll put it on my list to do or to follow up with mrhillsman about. | 20:01 |
| smcginnis | Don't have the brain space to do much with it at the moment until I clear up some other things. | 20:02 |
| *** david-lyle is now known as dklyle | 20:02 | |
| smcginnis | Oh webex... | 20:02 |
| ttx | ... | 20:05 |
| *** jamesmcarthur has joined #openstack-tc | 20:26 | |
| cdent | i had to give up on the board meeting, my brain won't do it | 20:29 |
| notmyname | I have another meeting at the same time. I'm hoping there's a recording. or at least detailed notes | 20:30 |
| dhellmann | they've said they will be sending the foundation staff slides to the mailing list | 20:31 |
| *** cdent has quit IRC | 20:39 | |
| * zaneb gave up because Java | 20:46 | |
| clarkb | zaneb: the POTS dial in number works | 20:47 |
| clarkb | https://docs.google.com/presentation/d/10UyCpxkjPqC3kT-dYRpBxzNT39i2OhlggJvzGDosMz0/edit?usp=sharing is the slides | 20:47 |
| zaneb | ah, I didn't see that | 20:47 |
| clarkb | I'm on POTS because it is easier | 20:48 |
| clarkb | but understand that toll free to the usa is not so easy for everyone :/ | 20:48 |
| zaneb | luckily it is for me | 20:49 |
| clarkb | I think we are on slide 46 now | 20:50 |
| *** diablo_rojo has quit IRC | 21:07 | |
| *** gouthamr_ has joined #openstack-tc | 21:18 | |
| mnaser | interesting read: | 22:07 |
| mnaser | https://lists.cncf.io/g/cncf-toc/message/2402 + https://lists.cncf.io/g/cncf-toc/message/2403 | 22:07 |
| mnaser | still an early thread | 22:07 |
| openstackgerrit | Alex Schultz proposed openstack/governance master: Add ansible-role-chrony to TripleO https://review.openstack.org/603516 | 22:08 |
| openstackgerrit | Alex Schultz proposed openstack/governance master: Add ansible-role-chrony to TripleO https://review.openstack.org/603516 | 22:11 |
| mwhahaha | jaosorior: -^ fyi | 22:11 |
| *** ricolin has quit IRC | 22:13 | |
| *** annabelleB has quit IRC | 22:36 | |
| *** diablo_rojo has joined #openstack-tc | 22:36 | |
| *** annabelleB has joined #openstack-tc | 22:37 | |
| *** jaosorior_ has joined #openstack-tc | 22:42 | |
| *** jaosorior has quit IRC | 22:45 | |
| mrhillsman | openstack-tc ML is moderated -_- | 22:47 |
| mrhillsman | how am i supposed to bug all the tc folks, bah | 22:47 |
| clarkb | mrhillsman: email the dev list? | 22:52 |
| mrhillsman | well, that makes me wonder why there is a tc list :) | 22:52 |
| mrhillsman | granted i do not think i have ever used it, or not even sure if it is used, btw, i do not care about the moderation hehe, figure it will get approved since it is not spam | 22:54 |
| dhellmann | mrhillsman : we use the tc list for meeting announcements and stuff, but try to keep all discussion on the open list | 22:54 |
| mrhillsman | i sent it to tc ML because it is specific to tc; discussion of proposal for joint meeting with uc on occassion | 22:55 |
| dhellmann | yeah, I think looking at the subject it's a good candidate for the public list | 22:56 |
| dhellmann | we're pretty aggressive about not using that private list :-) | 22:56 |
| mrhillsman | yeah i figured hehe | 22:56 |
| *** annabelleB has quit IRC | 22:56 | |
| dhellmann | I'll go ahead and discard this one, then? | 22:56 |
| mrhillsman | ugh, i have to write another | 22:57 |
| dhellmann | I rejected it so you should get a copy back ;-) | 22:57 |
| mrhillsman | 0.0 | 22:57 |
| dhellmann | or I could reply to the one that made it through to the uc list | 22:58 |
| mrhillsman | no copy just a you have been pwned | 22:58 |
| *** mriedem has quit IRC | 22:59 | |
| mrhillsman | i just figured it made sense to go to -tc based on the context and not -dev to possibly get missed | 22:59 |
| mrhillsman | thx for the reply | 22:59 |
| * dhellmann adds another example to the "why do we need to merge the mailing lists" discussion | 23:00 | |
| *** jamesmcarthur has quit IRC | 23:00 | |
| mrhillsman | ++ | 23:00 |
| *** jamesmcarthur has joined #openstack-tc | 23:01 | |
| *** jamesmcarthur has quit IRC | 23:07 | |
| persia | Is -tc a candidate for the merge? | 23:15 |
| *** tosky has quit IRC | 23:18 | |
| fungi | i could be convinced we don't really benefit from a separate -tc list. we already direct discussion to the -dev ml anyway | 23:49 |
| fungi | i feel like i lack sufficient background on why it was needed | 23:49 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!