15:00:03 #startmeeting FESCO (2018-10-08) 15:00:03 Meeting started Mon Oct 8 15:00:03 2018 UTC. 15:00:03 This meeting is logged and archived in a public location. 15:00:03 The chair is jforbes. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:03 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00:03 The meeting name has been set to 'fesco_(2018-10-08)' 15:00:03 #meetingname fesco 15:00:03 The meeting name has been set to 'fesco' 15:00:03 #chair nirik, maxamillion, jsmith, jforbes, zbyszek, tyll, sgallagh, contyk, bowlofeggs 15:00:03 #topic init process 15:00:03 Current chairs: bowlofeggs contyk jforbes jsmith maxamillion nirik sgallagh tyll zbyszek 15:00:06 o/ 15:00:09 .hello2 15:00:10 sgallagh: sgallagh 'Stephen Gallagher' 15:00:21 .hello psabata 15:00:22 contyk: psabata 'Petr Šabata' 15:00:32 .hello2 15:00:33 bowlofeggs: bowlofeggs 'Randy Barlow' 15:00:42 morning 15:01:29 .hello2 15:01:30 maxamillion: maxamillion 'Adam Miller' 15:02:18 Good, we have quorum 15:03:00 #topic #1974 Problematic blocker for F29: dnf 'offline' module tracking 15:03:00 .fesco 1974 15:03:01 https://pagure.io/fesco/issue/1974 15:03:03 jforbes: Issue #1974: Problematic blocker for F29: dnf 'offline' module tracking - fesco - Pagure.io - https://pagure.io/fesco/issue/1974 15:03:15 The ticket was updated on this one 15:03:53 I believe we have some agreement with the DNF regarding what needs to be implemented 15:04:03 with the required feature set for GA being pretty minimal 15:04:18 but I don't think there have been any technical development since last week 15:04:27 Hi, sorry for being late. 15:04:29 just discussions 15:04:33 freeze is tomorrow. ;( 15:04:40 So, it seems to me from the update, we get most of what is needed, with a few finer details to be added post F29 15:04:44 or well, 00:00 tonight 15:05:02 contyk: so what is the confidence that this code can be finished reasonably quickly? 15:05:14 within the next few hours? zero 15:05:52 Figured that, but do we have a real estimate? 15:06:34 I don't have any information on that matter, unfortunately 15:07:24 well, we can make this a fesco special blocker and hope... 15:07:44 Yeah, I don't think we really have much choice 15:07:45 I don't have a lot of confidence in that plan 15:08:09 well, the alternative is shipping what we have... 15:08:14 I mean, it would be important information to have, from a blocker stand point, are we talking about possible to get in and tested this week and not delay the release? Are we looking at 4 weeks of code and a guaranteed delay even if everything works perfectly? 15:08:17 iirc, we did seem to all agree that what we have is not great 15:08:38 jforbes: +1 15:08:52 yeah i agree - we don't have enough info to make a decision about this still 15:08:57 And what would be the risk of dnf as a 0 day update? 15:09:02 and last week we deferred because we wanted more info haha 15:09:12 well we did get more, but did not get the critical thing (an estimate0 15:09:48 bowlofeggs: the current agreement is progress, but yeah, an estimate is critical 15:10:02 Is there anyone who could provide such an estimate? 15:10:09 Well, what difference does it make, honestly? 15:10:27 sgallagh: the difference might be whether we ship what we have or mark this bug a blocker? 15:10:29 The current status is unsafe to ship with and reverting is likely more risky than fixing this bug. 15:10:33 jforbes: the dnf folks could but they aren't around atm 15:10:49 Our job is determining what constitutes being worth blocking on 15:10:57 sgallagh: how much risk would the fix being a 0 day update be? 15:10:58 sgallagh: yeah i guess it's a bit unsafe, though i wouldn't call it highly unsafe 15:11:31 most users aren't going to enable updates-testing repos unless they know a little bit about what they are doing (i don't claim they know a lot, but more than a pure GUI user would) 15:12:10 bowlofeggs: unless someone tells them to test a fix for something from updates-testing and they copy/paste commands from IRC 15:12:16 sgallagh: not knowing an estimate, do you advocate for holding the release for this bug even if it's a long time? 15:12:31 jforbes: true 15:12:57 though the pure GUI user probably won't find IRC either (maybe someone on a forum would tell them that though) 15:13:35 yeah it's not great if we ship what we have, that's for sure 15:13:35 "We believe a basic variant of this feature could be implemented for F29 GA" is the closest thing to an actual estimate in the bz update 15:14:08 actually gnome-software shows you repos now and lets you disable/enable them... 15:14:24 one option we have is to say this blocks, but also leave the door open to change our minds later? 15:14:29 bowlofeggs: I had to think about it a bit, but yes. I guess I am advocating for that, yes. 15:14:44 sgallagh: what about the 0 day update option? 15:14:44 nirik: oh wow - so the GUI lets you enable testing repos? i didn't notice that 15:14:52 if you install modular content and then decide to "disable modularity", as some people do, you will face the same problem 15:14:57 bowlofeggs: +1 15:15:01 nirik: heh, so it does 15:15:05 that's kinda neat actually 15:15:12 well that does make this more important 15:15:16 jforbes: I'm mostly concerned because once you get yourself into a bad state, it can't be fixed with an update. 15:15:18 there's a pointy-clicky way to end up in trouble 15:15:27 You need to do manual fiddly things to get back to the right place 15:15:41 sgallagh: fair. 15:16:16 proposal: file fesco special final blocker on this, try and get more estimate input from dnf team 15:16:21 Okay, given the information we have, since I don't see us getting any more information today, we we ready to vote on the blocker status? 15:16:34 nirik: +1 blocker 15:16:34 nirik: +1 15:16:38 nirik: +1 blocker 15:16:41 +1 blocker I fear 15:16:45 nirik: +1 15:16:46 I can take an action to pester them in person tomorrow 15:16:50 jforbes: I don't think we gain anything by waiting. 15:16:53 +1 to blocker 15:17:00 +1 to pestering from contyk ☺ 15:17:12 contyk: +1 15:17:22 * contyk takes a note 15:18:19 maxamillion ? 15:18:24 I don't think the MVP should take too long, it feels pretty trivial 15:18:29 +1 15:18:31 but we'll see what the estimate is 15:18:50 #agreed file fesco special final blocker on this, try and get more estimate input from dnf team (+7,0,-0) 15:18:53 I don't like any of this but I'm not really sure I have a better option ... :/ 15:19:05 #topic Next week's chair 15:19:14 who shall file the blocker? I guess I can if no one else wants to... 15:19:42 #info https://fedoraproject.org/wiki/Fedora_29_Beta_Release_Criteria#FESCo_blocker_bugs 15:19:50 I can file the blocker 15:20:05 #action jforbes will file the blocker 15:20:28 any takers for next week's chair? 15:20:37 sure 15:20:42 can do 15:21:08 Is it just a matter of marking 1616167 as blocking F29BetaBlocker ? 15:21:11 #action contyk will chair next meeting 15:21:27 #topic Open Floor 15:21:38 zbyszek: yeah, that would work 15:21:45 * bowlofeggs looks to the right 15:21:48 * bowlofeggs looks to the left 15:21:52 wow, it's so open in here 15:21:58 look under your feet 15:22:03 (for the crickets) 15:22:04 here I am ... stuck in the middle with you 15:22:12 :D 15:22:24 * sgallagh looks down and sees open air. He falls like Wile E. Coyote into the abyss 15:22:52 haha 15:23:10 sgallagh: that was neat how you held up the sign that said "help" and blinked twice before falling 15:23:12 I only said the floor was open, the walls are solid, this is a 6x6 cell 15:23:17 Hm, not F29BetaBlocker, F29FinalBlocker instead? 15:23:17 hahaha 15:23:44 yes, that 15:24:11 Actually, it is already a FinalBlocker, so I will just comment in it 15:24:30 Last call for open floor, going to close out in 1 minute 15:26:01 #endmeeting