18:00:08 #startmeeting FESCO (2015-09-02) 18:00:08 Meeting started Wed Sep 2 18:00:08 2015 UTC. The chair is jwb. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:00:08 Useful Commands: #action #agreed #halp #info #idea #link #topic. 18:00:08 #meetingname fesco 18:00:08 The meeting name has been set to 'fesco' 18:00:08 #chair ajax dgilmore hguemar jwb nirik paragan rishi thozza sgallagh 18:00:08 Current chairs: ajax dgilmore hguemar jwb nirik paragan rishi sgallagh thozza 18:00:09 #topic init process 18:00:15 .hello sgallagh 18:00:16 sgallagh: sgallagh 'Stephen Gallagher' 18:00:40 ajax will be missing today 18:00:53 Hey! 18:00:54 .hello hguemar 18:00:55 Hi all 18:00:55 number80: hguemar 'Haïkel Guémar' 18:01:10 * rishi tries the hello thing 18:01:12 .hello rishi 18:01:15 rishi: rishi 'Debarshi Ray' 18:01:27 .hello jkurik 18:01:28 jkurik: jkurik 'Jan Kurik' 18:01:31 hi all 18:02:10 i will wait 2 min for nirik, hguemar, and dgilmore 18:02:27 jwb: hguemar is here :) 18:02:29 jwb: I'm already there :) 18:02:33 .fasinfo hguemar 18:02:34 number80: User: hguemar, Name: Haïkel Guémar, email: karlthered@gmail.com, Creation: 2006-07-18, IRC Nick: number80, Timezone: Europe/Paris, Locale: en, GPG key ID: 26613DF3, Status: active 18:02:36 oh, lols 18:02:36 * kushal is visiting as usual :) 18:02:37 number80: Approved Groups: cla_fedora cla_done fedorabugs ambassadors cla_fpca gitbeefymiracle +packager python-sig provenpackager 18:02:59 yeah, i knew that. missed it somehow 18:03:00 sorry 18:03:06 I have 3 meetings in a row on wednesday 18:03:11 np :) 18:03:16 number80 you are the rockstar :) 18:03:19 you definitely lose. 18:03:27 hehe 18:03:40 sorry I'm late. ;) 18:03:52 reat 18:03:54 ok, moving on 18:04:00 #topic #1427 List of release blocking deliverables 18:04:00 .fesco 1427 18:04:00 https://fedorahosted.org/fesco/ticket/1427 18:04:03 jwb: #1427 (List of release blocking deliverables) – FESCo - https://fedorahosted.org/fesco/ticket/1427 18:04:15 i believe jkurik got feedback from most of the WGs 18:04:31 I summarize it in the last comment 18:04:55 so we seem to have the list. now what? 18:05:49 suggestion: make the fedora 24 list now also... 18:06:08 i'm not opposed to that, but we should do something with the list we have first... 18:06:09 for the future we need to think of a way how to have this in a machine readable form 18:06:20 like... give it to rel-eng and QA? 18:06:21 also, the list seems to be missing the atomic stuff? 18:06:32 nirik, the cloud people said no atomic 18:06:34 and vagrant images 18:06:42 no atomic at all? we are dropping it? 18:06:48 nirik: it's a spin 18:06:54 not a release blocker 18:06:54 the atomis stuff is not specifically related to a Fedora release 18:06:55 which is really weird, because they also said atomic is going to be the primary focus of the cloud WG 18:06:59 so... 18:07:14 nirik, i think it's because of the 2 week release cycle thing 18:07:22 hum, ok thats confusing, but ok. 18:07:25 Atomic is specific 18:07:27 perhaps a note on there about that? 18:07:42 nirik: ok, will put a note into the wiki 18:07:47 also I think we are missing arm images? 18:08:00 where? 18:08:11 Workstation never wanted arm images 18:08:12 http://dl.fedoraproject.org/pub/alt/stage/23_Beta_TC1/Images/armhfp/ 18:08:27 I guess thats just another arch for spins 18:08:43 so Server is missing it i guess 18:08:49 arm is mentioned here: https://fedoraproject.org/wiki/Fedora_Program_Management/ReleaseBlocking/Fedora23#Non_Release_Blocking_Deliverables_.28pls_fix_if_it.27s_blocking.29 18:09:07 jkurik: thats the server arm image. 18:09:13 the ones I pointed to are spins 18:09:19 ah 18:09:32 http://dl.fedoraproject.org/pub/alt/stage/23_Beta_TC1/Docker/x86_64/ also seems not there. 18:09:39 nirik, we don't block releases for spins 18:09:53 so how does that fall into this ticket? 18:10:03 well, this is all deliverables I thought. 18:10:11 if it's not on the page we should not be making it. 18:10:12 release blocking deliverables 18:10:19 the ticket says release blocking 18:10:22 no? we mention lots of non release blocking there? 18:10:34 dude, i'm going by the ticket. 18:10:38 Nitpick. It is a bit confusing ot have "non release blocking deliverables" in a page that says "release blocking". 18:10:41 "List of release blocking deliverables" 18:11:00 the orig ticket is a bit confusing 18:11:11 "release engineering knows what's going to be produced for upcoming release ahead of time (we added this point to the Change process as releng requested it)" 18:11:22 well, that would be everything we produce, not just release blocking... but ok. 18:11:56 I was thinking this was all deliverables and marking what was release blocking... 18:12:01 ok, so is there anything further to do with the list itself? 18:12:09 because if somethign is missing, that's up to rel-eng and the WG to sort out 18:12:10 (back to the atomic thing, it will become cloud WG primary deliverable starting F24) 18:12:20 if it makes sense for you 18:12:40 number80, not really, since it still won't be tied to a fedora release 18:12:57 how can you ahve a primary deliverable for a release when it is released every 2 weeks? 18:13:12 is the one that happens to land close to the F24 release the "release deliverable"? 18:13:19 i'd rather sort that out later though 18:13:28 so, yeah, send this on to releng and qa I'd say... 18:13:33 jwb: ack, but I agree that we need to clarify this 18:13:39 I can try and add non release blocking stuff I see that we make. 18:13:48 from the ticket point of view I believe we have all blocking deliverables listed; just the non-blocking might want some update 18:13:54 perhaps devel announce to make sure people shout if their deliverable is missing? 18:14:12 jwb, does everything has to follow the same release story? I means the 6 months story. 18:14:16 s/means/mean 18:14:35 proposal: Send current release blocking deliverable to rel-eng, QA, and devel-announce 18:14:57 jwb: +1 18:14:59 jwb: sounds good to me +1 18:14:59 kushal, no, but it is very confusing to talk about something that is following it's own release story in the context of a story it isn't part of 18:15:06 jwb: +1 18:15:07 jwb: +1 18:15:11 +1 18:15:14 jwb, yup, understood. 18:15:43 +1 18:16:00 #agreed Send current release blocking deliverable to rel-eng, QA, and devel-announce (+6, 0, -0) 18:16:11 ok, so next steps 18:16:14 actually hold 18:16:18 who is doing the sending? ;) 18:16:19 jkurik, can you do that sending? 18:16:35 jwb: Yes I can and I will 18:16:37 thank you 18:16:42 #info jkurik to send things around 18:16:46 ok, next steps 18:16:50 1) generate the f24 list 18:16:56 2) sort out what Atomic means 18:17:01 anything else? 18:17:50 jwb: machine readable format ? 18:18:12 releng was interested in this 18:18:13 sure 18:18:35 #info Next steps are to look at a machine readable format, generate the f24 list, and sort out how Atomic plays into this 18:18:43 ok, anything else? 18:19:14 moving on 18:19:20 #topic #1444 updates deliverables 18:19:20 .fesco 1444 18:19:20 https://fedorahosted.org/fesco/ticket/1444 18:19:21 #topic #1444 updates deliverables 18:19:21 jwb: #1444 (updates deliverables) – FESCo - https://fedorahosted.org/fesco/ticket/1444 18:19:32 i don't think we've seriously thought about this in one week 18:19:41 particularly since nobody commented the ticket 18:19:45 defer? 18:19:49 +1 18:20:36 yes defer +1 18:20:48 +1 to defer 18:20:50 +1 18:21:05 Yes, lets defer. (also dgilmore is missing today) 18:21:18 #agreed Defer this topic for now (+6, 0, -0) 18:21:23 #topic #1467 F23 Changes - Progress at Change Checkpoint: Completion deadline 18:21:26 .fesco 1467 18:21:28 jwb: #1467 (F23 Changes - Progress at Change Checkpoint: Completion deadline (testable)) – FESCo - https://fedorahosted.org/fesco/ticket/1467 18:21:29 https://fedorahosted.org/fesco/ticket/1467 18:22:13 there was supposed to be some update, right? 18:22:26 yes. 18:22:48 passphrase policy should be all set (anaconda changes should be in) 18:22:59 on dnf upgrades the package was reviewed and approved and is now in 18:23:30 the rest is still in ASSIGNED 18:23:55 networkd is deferred to F24 18:24:01 why isn't the dnf bug closed? 18:24:40 which one? 18:24:47 the review is waiting on git creation. 18:24:55 * nirik can go do that now 18:25:04 you said it was "in" so i assumed that had already happened 18:25:10 I think the bugs should be moved to MODIFIED / CLOSED or deferred 18:25:17 looks like the lorax changes to drop updates.img is complete 18:25:27 only DNF bug has some comment after jkurik's reminder 18:26:06 change bugs shoudl be in ON_QA now 18:26:46 nirik: not now but the next week 18:27:08 http://red.ht/1Us6OQv - this is the current list of tracking bugs 18:27:24 ok. 18:27:43 I don't know the status of 2 week atomic aside that there's a bunch of work on it... so hopefully it's close to ready. 18:27:50 number80, ? 18:28:16 i recall seeing announcements about the current images being spit out, so hopefully it's done 18:28:20 jwb: i know that releng are working on it but didn't get updated status 18:28:33 Layered Docker Image Build Service (https://bugzilla.redhat.com/show_bug.cgi?id=1243736) seems a bit too late for F23. 18:28:53 yeah, we're probably going to have to defer that to f24 at least 18:29:11 I can prod folks to update status on those. 18:29:19 The bug is basically empty, and I can't see any other bugs being tracked. 18:29:20 "Layered Docker Image Build Service" and "Two Week Atomic" are not strictly related to F23 18:29:38 right, it can happen asynchronously 18:29:40 the first one should be 18:29:45 then they should have no problems being deferred to another release they aren't strictly related to 18:29:51 I am going to talk to owners of these two Changes how to deal with it 18:30:29 ok, so it sounds like 2 of the 4 left are basically done. one of them is unknown (layered docker) and the other is already well underway (2 week atomic) 18:30:44 right 18:30:50 does anyone want to do anything specific in this meeting with these then? 18:31:36 well, let's have jkurik reach out owners and emit final decision next week for those @ 18:31:38 2 18:31:48 I think we should comment in those that if these are not updated and moved to another state till next week, they are deferred 18:31:56 we can't wait forever 18:32:06 thozza, i'm fine with that. anyone disagree? 18:32:14 fine with me. 18:32:16 I thing we can wait for the deadline for Change Checkpoint: 100% Code Complete Deadline - 2015-Sep-08 and then sort it out 18:32:45 jkurik, that's a day before the next meeting 18:32:51 yes 18:32:58 thozza, would you be able to make those comments? 18:33:05 jwb: sure 18:33:34 #info thozza to comment in bugs without update and we will revisit next week 18:33:39 anything else on this? 18:34:16 looks like we're done with this 18:34:23 * nirik has nothing more on this 18:34:24 I already made a coments; i.e.:https://bugzilla.redhat.com/show_bug.cgi?id=1215939#c2 18:34:27 ok. sgallagh, who i thought was here but hasn't said anything at all yet, asked us to skip the systemd presets ticket so moving on to new business 18:34:43 jkurik, more can't hurt ;) 18:34:44 #topic #1473 Changes approval process should reflect the existence of WGs 18:34:47 .fesco 1473 18:34:48 jwb: #1473 (Changes approval process should reflect the existence of WGs) – FESCo - https://fedorahosted.org/fesco/ticket/1473 18:34:50 https://fedorahosted.org/fesco/ticket/1473 18:34:57 this came from a discussion at flock 18:35:08 has everyone had time to review the proposal? 18:35:11 I'm ok with this, but it sometimes might seem heavy weight... 18:35:53 yeah, especially as WG should have liaison with fesco 18:35:53 right. It was more of an idea. I'm open to suggestions. But I think the existence of WGs should be reflected in the process in some form 18:36:22 number80, that's actually a good point. the liasions used to be here every meeting. that was the primary purpose for them 18:36:36 to interact between FESCo and the WGs 18:36:48 stickster is here 99% of the time 18:36:56 sgallagh is the liasion and is normally here 18:37:01 I am fine with it, however the mechanism of getting approval from WG is not clear for me. 18:37:02 the only thing I could think of was to add some kind of lazy consensus... ie, if a WG does not give feedback, assume it's ok to move on... but not sure how to codify that without making it more complex. 18:37:05 /me waves 18:37:09 thozza, If Change says it needs to work with Fedora Council or Fedora releng or Fedora qa then should tickets be also filed in their ticketing system? Or FESCo will decide it later based on WG's response? 18:37:28 jkurik: the problem is that some WGs don't have ticketing system 18:37:35 which don't? 18:37:41 thozza: yes, I know, that is the problem 18:37:45 jwb: Server doesn't really have one. 18:37:51 jwb: what's up? 18:37:53 paragan: I think they should be filled ideally 18:38:01 sgallagh, you have a trac instance, don't you? 18:38:05 We sort of unofficially use the fedora-productimg-server BZ for handling edition-level bugs 18:38:05 should not have every WG a ticketing system ? 18:38:14 maxamillion, we've passed by your topic now. i'll chat in #fedora-devel 18:38:31 jwb: alright 18:38:32 e.g. the Workstation WG has track linked on their wiki 18:38:37 jkurik: that was left to their appreciation, never was a requirement 18:38:39 Server WG has only a blog 18:38:40 We tinkered with using Trello/Cantas, but it didn't go anwhere 18:40:05 as long as there is a way to reach a working group (they all have lists), I'm fine with tickets being in fesco track ccing their list 18:40:18 I think we should have a unified way across all WG how to get their approval 18:40:28 nirik: that was my idea 18:40:33 nirik: +1 for centralizing on fesco ticketing system 18:40:57 going back and forth between trac instances won't make it easy 18:41:20 thats all technical details, we can figure something out. 18:41:26 number80: my idea was that the discussion with WG can get really extensive 18:41:42 but it may be OK to have it all in one ticker 18:41:54 thozza: I think that's the role of liaison to organize the discussion 18:41:59 also FESCo members will be then more aware of the possible issues 18:42:16 number80: but anyone can comment in the ticket 18:42:22 or just fesco ticket and someone mails that to the working group list. 18:42:36 nirik: also an option 18:43:02 thozza: yes, if WG wants to have their own ticket for internal discussion, their prerogative 18:43:04 and having the WG liaison to comment in the ticket for the whole WG 18:43:42 number80: right. my comments were about using only FESCo ticket also for WGs discussion 18:44:06 thozza: that is my prefference as well 18:44:30 thozza: we could, but they have very different organisations 18:44:36 I think we don't have to decide the ticket today, but we could at least add some ideas to the ticket. I can update the proposed change little bit 18:44:40 * rishi got pulled away into a real-life conversation 18:44:49 What is this "TL;DR I am against ..." ? 18:45:06 thozza: globally, i'm +0.5 for the current proposal 18:45:13 rishi, people don't like more process 18:45:18 shocking 18:45:21 :D 18:46:59 so, tune proposal, revisit next week then/? 18:47:20 As I was vaguely saying the other day, I think instead of liaisons, we should make them full fledged FESCo members. 18:47:31 ie. ask each WG to nominate someone to FESCo. 18:47:39 my preference is to use FESCo trac with WG mailing lists on CC and apply lazy consesus if a WG does not react 18:47:41 And have a few non-WG members too, of course. 18:48:08 rishi, that wasn't done originally because the _primary_ focus for a liasion should be the WG stuff 18:48:20 not FESCo with WG on the side or vice versa 18:48:41 rishi, also: that is a huge conversation that isn't really related to this ticket and isn't going to be solved over IRC 18:48:58 so... maybe draft a proposal for re-seating FESCo and send it to the list if you'd like to pursue that 18:49:16 jwb: Well, I doubt there are too many people who know everything that goes into Fedora in the same level of detail. 18:49:26 rishi: personally I'm against "reserving" any FESCo seats for WGs. They can nominate anyone and try... 18:49:35 rishi, great. draft proposal, send to list. 18:49:45 well, I don't see the necessity to have liaison full fledged members of fesco as we are a consensus based group 18:49:48 let's not derail this meeting at the moment 18:50:16 Well, this ticket is about reflecting the existence of WGs. 18:50:58 then, let's follow nirik proposal and continue the discussion on trac 18:51:29 number80: sounds good to me 18:51:31 number80, +1 18:51:34 Ok, +1 18:51:48 yes, trac. +1 18:52:13 i'm going to assume nirik is ok with his own suggest 18:52:17 yes. +1 18:52:31 #agreed Continue discussion in trac (+5, 0, -0) 18:52:41 #topic Next week's Chair 18:52:45 who wants it? 18:53:16 #info jwb to chair next week 18:53:19 #topic Open Floor 18:53:32 anything for Open Floor? 18:53:42 * nirik has nothing. 18:53:46 jwb: You are too fast. I could have done it to atone for my network fuck up last week. 18:54:21 rishi: then you have the week after that 18:54:48 ok 18:55:56 anything for Open Floor? 18:56:05 * jkurik has nothing 18:56:21 ok. ending the meeting in 55 seconds 18:57:26 #endmeeting