20:03:22 #startmeeting tc 20:03:23 Meeting started Tue Jan 26 20:03:22 2016 UTC and is due to finish in 60 minutes. The chair is flaper87. Information about MeetBot at http://wiki.debian.org/MeetBot. 20:03:24 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 20:03:26 * devananda lurks 20:03:26 The meeting name has been set to 'tc' 20:03:34 #topic Agenda 20:03:36 #link https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee 20:03:42 you can find our agenda for today there 20:03:52 #topic remove expired extra ATCs 20:03:54 #link https://review.openstack.org/268272 20:04:00 #chair dhellmann 20:04:01 Current chairs: dhellmann flaper87 20:04:04 (in case my network drops) 20:04:17 the meeting today should be pretty straightforward 20:04:31 This has 6 votes 20:04:32 so this is some cleanup I did in the course of changing the yaml schema to include extra-atcs for fungi 20:04:35 wow i forgot about #chair 20:04:38 I reached out to the people on the docs list, some of those emails even bounce, but generally people are not surprised 20:04:52 ok, good 20:05:12 it's not like it's hard to add folks back to the list, either, if the ptl wants 20:05:14 dtroyer: your vote is missing on that one :d 20:05:19 dhellmann: right 20:05:35 what I wondered was whether those names were included in the round of Austin summit invites? 20:05:45 agentleweb: I don't think they were 20:05:48 but I'm not 100% sure 20:05:54 if they're expired and they were included, they got lucky :) 20:06:02 russellb: my thoughts exactly 20:06:07 heh 20:06:07 flaper87: done 20:06:13 but of course, easy to re-add if applicable and needed 20:06:40 I don't have the rights to approve that, just ttx does. We won't be approving anything today, we'll let that for ttx to do it when he's back from the beach 20:06:47 sounds good 20:06:54 Let's try to at least get some consensus on stuff 20:06:58 anything else on this topic? 20:07:19 moving on 20:07:21 we should have a CI job that automatically checks for expirations and proposes these changes 20:07:23 #topic Applying for a EC2API replacement project. 20:07:23 (kidding) 20:07:24 #link https://review.openstack.org/268774 20:07:46 russellb: I believed you for 1 second 20:07:48 :P 20:08:12 this one has some comments from agentleweb 20:08:27 but I don't think there's any major objection 20:08:32 russellb: we actually do that in infra ;) 20:08:33 All I did was look up EC2 trademark info, and then I think the mission could be clarified is all 20:09:15 agentleweb: something that could be done in a follow-up patch? or you'd rather have it in this one 20:09:17 ? 20:09:18 agentleweb : good point on the wording, see my follow-up 20:09:29 we could call it the FD3 API if needed 20:09:38 * flaper87 is still refreshing the browser to see dhellmann's comments 20:09:55 why not DA2? :P 20:10:08 lol 20:10:26 ok, this one needs some extra work 20:10:48 look forward to the new PS and vote again :D 20:11:20 just minor wording and I believe we're good. 20:11:25 anything else? 20:11:49 ooooooooook, moving on 20:11:52 #topic add new Repo(shovel) to the Governance Repository 20:11:54 #link https://review.openstack.org/269417 20:12:12 dhellmann: and ttx dropped some comments there 20:12:26 I didn't commented/voted because I very much agree with their comments 20:12:38 I probably should've added my vote there as well 20:12:42 same, i didn't dig much with the existing -1s 20:12:52 we probably don't need to pile on, if folks agree 20:13:01 can I hear more about rackHD or get some reading material? 20:13:06 baremetal driver? 20:13:18 agentleweb : https://rackhd.readthedocs.org 20:13:42 I didn't dig much into rackhd itself but based on the maturity of the project (they literally just announced it) I think it'd be better to wait 20:14:06 I wonder if the request for inclusion in the tent was made on the wrong assumption that it is arequirement for having a repo 20:14:09 jroll: do you have any insight into this project? 20:14:12 thanks dhellmann 20:14:16 from what I read, it's a single vendor backed solution 20:14:18 ohai 20:14:27 devananda : yeah, that was my impression, too 20:14:32 * devananda is not jroll, but was already typing that out 20:14:42 ah and devananda is lurking too :) 20:14:43 so, yeah, rackhd is brand new thing that is mostly an ironic competitor with a small featureset that does not overlap 20:15:00 shovel is even newer, and basically a bridge to pick up nodes from shovel and put them in ironic 20:15:09 and push extra monitoring data into ironic db 20:15:25 i think i understand the name now 20:15:31 I personally would prefer that EMC contributed to ironic, but they don't seem willing to do so 20:15:35 do either shovel or rackhd have features that were on the road-map for ironic? 20:15:52 jroll: have they explained why? 20:15:56 dhellmann: rackhd has features we'd like to do, yes 20:16:06 jroll: I mean, why they don't contribute to Ironic 20:16:08 jroll : thanks 20:16:25 flaper87: because they would rather rackhd not be an openstack thing, is all I've really heard 20:16:38 am I remembering correctly that either shovel or rackhd or both were written in node? 20:16:42 both 20:16:47 and yet there's a library that sits on top of it that aims to be part of openstack 20:16:48 ok 20:17:00 sounds fishy. 20:17:01 flaper87: yep, they want to integrate with openstack, but not require openstack 20:17:20 it's essentially an ironic-inspector replacement, at that point, best I can tell 20:17:25 is keedya in channel today? 20:17:37 tab complete says no (his nick is keedya 20:17:39 ) 20:17:39 the description here sounds very similar to ironic ... https://rackhd.readthedocs.org/en/latest/how_it_works.html#general-bare-metal-automation-with-pxe 20:17:56 and the next section on Discovery -- also duplicates what we do with Inspector 20:17:59 We've rejected requests because of the language they are built in. There are a couple of more reasons in the case of Shovel besides the language thing 20:18:10 jeblair: can we support node in our current infra? 20:18:33 dhellmann: to you rlanguage question, from the docs: "The code for RackHD is a combination of JavaScript and C" 20:18:57 duplicating features is no longer forbidden outright, so the question is whether this is "gratuitously competing or reinventing the wheel" 20:19:11 agentleweb: krotscheck has been doing much work around improving npm testing. i think javascript in general is either well supported or close enough to be well supported not to have concerns. however, i don't know if there are special challenges around node specifically. 20:19:12 devananda : cool, thanks 20:19:39 yeah, so it's weird, because it's rackhd that is reinventing the wheel 20:19:46 however only shovel is applying to be an openstack thing 20:20:01 yeah 20:20:12 from a governance perspective I don't think we're ready to accept projects written in other languages. Are we? I remember we discussed this like 3 or 4 months ago 20:20:32 agentleweb: lets say i should ask infra at large about node in particular. 20:20:32 "integrate with rackhd" isn't a thing I'm opposed to, I just wonder why it needs a whole project with a ptl and everything 20:20:41 Shovel is just a proxy to an ironic akin software 20:20:42 rather than building a thing and talking about joining the ironic team 20:20:56 or integrating it into ironic itself 20:21:00 jroll: is that something that could be part of Ironic's team? 20:21:03 or even project 20:21:07 jroll: yeah 20:21:12 sorry, you write faster than me 20:21:18 :) 20:21:20 jroll : I would be interested in hearing that discussion, too. It came up in the context of astara as well, but in a different way 20:21:27 sorry I'm late, ECHILD. 20:21:31 I'd be opposed to it today, but maybe not in the future 20:21:50 flaper87: our guideline concludes that as long as they can build on infra and docs and test in the gate another language is ok 20:21:58 jroll, others: has rackhd itself applied to join openstack -- or just shovel? 20:22:05 devananda: just shovel 20:22:07 agentleweb: a-ha! ok, we're good there 20:22:08 weird 20:22:12 devananda : this application is only for shovel 20:22:21 devananda: the repo is already openstack, it's just the governance part we're discussin 20:22:25 I do not understand why a shim on top of ironic would be a separate project 20:22:29 ok, I think we have enough questions to feed back to the review 20:22:43 jroll: ? shovel is openstack ? 20:22:44 And we need to sit this and wait for those answers 20:22:51 jroll: I am confuse. 20:22:59 lifeless: openstack/shovel exists, the governance patch is the discussion at hand 20:23:07 lifeless, jroll : https://review.openstack.org/#/c/269411/ 20:23:18 yeah that one. 20:23:19 In summary, it seems to me that the agreement is: As it is, Shovel is not ready to stand by itself unless it is pulled under the Ironic's team umbrella 20:23:23 does that sound fair? 20:23:26 jroll: its not an official project though, which is what I thought you were implying by saying it 'is openstack' 20:23:53 flaper87 : there are questions about that, yes. My concern is that the project is young, and the contributors haven't engaged with the ironic team at all. 20:23:59 lifeless: yeah, sorry, by "the repo" I only meant code 20:24:06 words are hard 20:24:33 dhellmann: agreed, lets feed all that back into the review and wait for feedback from the authors. 20:24:38 flaper87 : I'm not sure we should assume that the project must become part of ironic 20:24:48 dhellmann: agree 20:24:53 dhellmann: didn't mean t oassume that rather propose as an option 20:24:59 k 20:25:15 dhellmann: ++ on both points 20:25:55 #action flaper87 to comment back on the Shovel review and provide feedback based on what was discussed here 20:26:00 anything else before we move on? 20:26:04 yeah, sounds like the right next thing for shovel 20:26:37 ok, moving on 20:26:41 #topic Adding SaltStack to OpenStack 20:26:43 #link https://review.openstack.org/269556 20:26:58 flaper87: I commented on the review 20:27:06 Same as for Shovel, didn't comment because I agree with the current comments 20:27:07 the somewhat inaccurately named patch :) 20:27:09 lifeless: oh, nice! thanks 20:27:31 jeblair: yeah, I was confused when I read the title 20:27:36 "OH Really?" 20:28:11 so, I don't think there are objections here other than maturity 20:28:19 That seems to be the feedback from the review 20:28:46 is cznewt in channel today? 20:29:17 doesn't look like it 20:29:56 mmh, doesn't look like 20:29:58 :/ 20:30:18 I'm a little surprised that there haven't been any responses to the review comments already posted on this one 20:31:07 also, the author of the review is a new contributor. I don't think that's bad but I believe that bakes up the feeling that the project probably needs some extra time 20:32:00 yeah, it was surprising that someone's first patch was to add a project (not bad, just surprising) 20:32:14 ok, will try to contact cznewt and get feedback 20:32:24 flaper87: +1 on it baking the feeling :) 20:32:36 usually the first patch is to create the project in project-config 20:32:37 anything else folks want to add ? 20:32:42 jeblair: right 20:32:43 :D 20:32:47 nom baked feels 20:33:11 ok, moving on 20:33:14 patch for project-config was submitted today 20:33:24 AJaeger : ah, good to know 20:33:30 honestly though, after adding a project recently, I can see why you'd sorta "ask first, do all the config work later" :) 20:33:33 AJaeger: oh, thanks for the heads up 20:33:45 and they had their first ever irc meeting 20:34:00 agentleweb : heh 20:34:05 Just like in Shovel's case, I wonder if this happened because ppl think being an official project is a requirement for having a project 20:34:07 And there was a second patch for project-config - seems the two teams are working together now ;) 20:34:31 nice thanks for the reporting AJaeger! 20:34:38 * agentleweb is anti-comma today 20:35:15 ok, I guess we can move on now, if there's nothing else to add here 20:35:35 #topic Adding SaltStack to OpenStack 20:35:36 #link https://review.openstack.org/269556 20:35:38 ops 20:35:45 #topic improve licensing docs 20:35:47 #link https://review.openstack.org/269823 20:36:29 agentleweb: and ttx have some comments on it 20:36:42 looks like minor changes required 20:37:14 dhellmann: is that something you can do now? That way we can vote right away. 20:37:17 those explanations are super helpful as are the edits, but I'm still not sure what the limiters to redistribution under CLA mean 20:37:24 * flaper87 didn't vote as there was going to be a new PS anyway 20:37:31 so muchas gracias for this! 20:37:35 agentleweb : I'm not sure what that question means? 20:38:02 dhellmann: I can't get this: "which allows redistribution under ASLv2 (currently limited to MIT and BSD (both forms)" 20:38:22 the CLA, it allows redistribution under ASLv2. then what about MIT and BSD forms? 20:38:39 i actually don't understand that at all 20:38:53 the cla grants the foundation the rights to distribute code under anything it wants 20:39:09 the bylaws take away the rights for the foundation to distribute it under anything but asl2 20:39:17 I believe the intent of that section is saying that we think the MIT and BSD licenses are compatible with the redistribution requirements of the CLA 20:40:29 I think adding in the actor as "The Foundation" may help... Foundation can redistribute under these three licenses: ASL2, MIT, BSD. 20:40:43 i don't think it can 20:40:49 jeblair: ah ok 20:40:57 I think agentleweb has it flipped 20:41:01 so why are MIT and BSD in there at all? 20:41:14 The CLA allows ASL2, MIT, BSD, then the Foundation can redistribute under ASL2 20:41:25 is my reading of the intent 20:41:25 the cla doesn't need to allow anything 20:41:34 ok, wording updated 20:41:55 ah, flipped, got it. 20:42:14 the part of the wording that says something should be licensed under a license compatible with the asl2 (which are mit and bsd) is correct. 20:42:19 i don't know what it means to be supported by the cla 20:42:47 jeblair: the bylaws only mention the approved release 20:42:57 jeblair : should that part be just "licensed under the CLA..."? 20:43:01 jeblair: no mention is made about things produced that are not in the approved release 20:43:12 oh, right, I think that's why "supported by" is in there 20:43:22 * flaper87 is following but he doesn't have much to add to the wording 20:43:28 jeblair: I think this is at best ambiguous 20:43:28 this is deep english we're discussing 20:43:30 :D 20:43:39 heh 20:43:48 well, it says "licensed under a license supported by the cla" 20:43:50 basically, projects have to decide right up front if they want to ever be in the approved release, and be careful to select a license that supports the further constraints 20:43:56 (our document says that ^) 20:43:58 jeblair: nevertheless, doug's patch doesn't add BSD and MIT, he is merely trying to fix the prose 20:44:00 legal language is the hardest language syntax there is 20:44:09 agentleweb: no kidding 20:44:12 flaper87 : legal language is not english 20:44:18 lifeless: i agree that the problem exists in the original text dhellmann is modifying 20:44:22 jeblair: I agree that there is an issue here, I think its orthogonal to the patch. 20:44:25 jeblair: ok, cool. 20:44:30 dhellmann: ah, I'll put that language in hte list of languages I should learn 20:44:31 or not 20:44:33 :P 20:44:42 it does make it hard to evaluate the patch though :| 20:44:43 the intent of this patch is to clearify the language, so if we can do that I'm happy to continue making edits right now 20:45:03 we still have some mins and this is the last topic 20:45:07 lets keep iterating 20:45:09 how about "must use a license supported by"? 20:45:10 basically, without getting to the root of that, i think all we can hope to do is make it 80% clear :) 20:45:27 I think keep editing, yah. 20:45:29 so I'm fine with improving the languagge; I think we need to actually revisit the original intent though; IIRC it was a wiki import? 20:45:49 the point, I think, is that if a project chooses the GPL up front they can never be a part of the approved release because the foundation can't release the code under the ASLv2, right? 20:45:58 and we were trying to say that more generically? 20:46:11 dhellmann: unless the authors of the code sign the cla, at which point they can 20:46:21 and I guess "never" is too strong, since theoretically they could relicense it 20:46:27 jeblair: I'm speculating: perhaps since this was about what things could be hosted on git.o.o, perhaps the intent was to say we couldn't host things that the cla didn't permit 20:46:33 jeblair : but can the foundation change the license just because of the CLA? 20:46:36 jeblair: since everything had to be under the clas 20:46:59 lifeless : no, that's not the intent 20:47:11 dhellmann: i believe so. i believe that if it weren't for the bylaws, the foundation could actually gpl all of openstack. 20:47:12 we clearly say later in the document that other licensees are ok 20:47:25 the only issue is a project that wants, someday, to be part of the approved release 20:47:33 dhellmann: later in the document isn't talking about what we produce 20:47:38 jeblair : ok, I am not sure the GPL terms allow that, that's why I ask 20:47:41 if that comes as a surprise to anyone who has signed the cla -- well, this is one of the reasons i don't like the cla. 20:48:02 jeblair: the surprise part or the signing part? :) 20:48:11 dhellmann: they don't need to. the cla literally allows the foundation to distribute code under any terms it wishes. its bylaws constrain it to doing so only under the asl2. 20:48:14 jeblair: yes, i believe you're correct 20:48:28 CLAs are often used to keep that option open 20:48:30 jeblair: that sounds right to me in my limited understanding :) 20:48:34 and technically it's still open if we changed the bylaws 20:48:47 isn't it still open for anythong outside the approved release ? 20:49:34 dhellmann: (considering the apache license basically lets you do that anyway, why we need the cla in that case is still a throw-hands-up-in-the-air puzzle :) 20:49:43 yep 20:49:49 lol 20:50:01 jeblair : yes, I agree 20:50:21 crinkle: I believe/agree that we should store/present that too if we make a thing 20:50:34 * dhellmann wonders what meeting mordred is in 20:50:39 lol 20:50:44 dhellmann: i think your latest clarification makes things as clear as possible 20:50:49 o/ 20:51:17 and the next step after that is to see if we can clarify what "supported by the cla" is intending 20:51:52 ok 20:52:02 ok, I'll re set my vote 20:52:08 we can either take this patch, or continue refining it, either is fine with me 20:52:21 dhellmann: I'm in "buffered lines sent after landing from airplane" meeting 20:52:24 lgtm 20:52:25 if folks here can do the same, it'd be cool. That way ttx can approve it when he's back 20:52:42 mordred : welcome to your future 20:52:53 * mordred marvels at the wonder 20:53:01 mordred : what an age to be alive 20:53:01 dhellmann: that clarifies along my understanding, thanks 20:53:10 ok, we need to move on now 20:53:31 dhellmann: I think keep refining it 20:53:49 we can move on, sorry flaper87 20:53:51 agentleweb : wfm 20:53:53 #topic Comms WG update 20:54:02 so, a post went out last week 20:54:22 and one explaining the changes w.r.t linux images and OpenStack is on the works 20:54:35 I'm hoping to get it done this week and I'll need reviews from folks 20:54:39 I have a TC blog post, it was done Friday! But, uh, I need someone to push a button and can't get in touch with Allison at the Foundation... 20:54:47 anyone else have push-button rights to the blog? 20:55:05 This turned out to be a crazy week for me so, I really hope I'll get it done 20:55:13 agentleweb : they're at their off-site this week, ttx may be able to reach them 20:55:15 agentleweb: ah, thought it went out 20:55:17 ok 20:55:32 flaper87: yeah, sadly, not quite 20:55:48 ok, it's done! That's good 20:55:55 yeah! 20:56:01 I don't think we need one for this meeting, though 20:56:05 We might need one next week 20:56:10 yeah 20:56:19 ok, I don't have anything else on this topic 20:56:22 agentleweb: yuo? 20:56:25 you,even? 20:56:25 any other comms- channels we should be paying attention to? 20:56:31 yo, yes! 20:56:33 :) 20:56:40 I think we're good 20:57:02 coolio! 20:57:06 #topic Open Discussion 20:57:15 we've 3 mins left 20:57:26 hope this meeting was good enough :D 20:57:30 * flaper87 admires ttx 20:57:35 thanks for stepping in, flaper87 20:57:42 my pleasure 20:57:44 thanks flaper87 20:58:04 * agentleweb hopes the summit proposals are coming in strong 20:58:06 flaper87: nicely done! 20:58:11 real quick, I'll write something for the project team guide re-stabilizatoin cycles 20:58:16 * flaper87 bows 20:58:25 flaper87: nice job! 20:58:35 I don't think many folks liked the idea but I still think we should document it 20:58:35 agentleweb : me, too, is there some way to see the proposals, yet? 20:58:48 dhellmann: not yet that I know of 20:59:02 in infra-land, folks are coordinating on proposals but have not submitted them yet 20:59:14 other than that, I think we won't know whether they are really good/bad until someone gives them a try 20:59:33 that way, we get the right co-presenters on early, so we don't have 50 people proposing 'intro to infra' 20:59:44 ok, we're out of time 20:59:47 jeblair : ++ 20:59:49 thanks everyone! 20:59:59 have a good one and tty all next week 21:00:04 #endmeeting