17:00:46 #startmeeting FESCO (2014-09-10) 17:00:46 Meeting started Wed Sep 10 17:00:46 2014 UTC. The chair is jwb. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:46 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:00:46 #meetingname fesco 17:00:46 #chair dgilmore jwb kalev mattdm mitr mmaslano nirik sgallagh t8m thozza 17:00:46 #topic init process 17:00:46 The meeting name has been set to 'fesco' 17:00:46 Current chairs: dgilmore jwb kalev mattdm mitr mmaslano nirik sgallagh t8m thozza 17:00:53 ok, let's get rolling 17:00:53 morning everyone. 17:00:56 who's around 17:01:02 hola 17:01:04 Hello 17:01:10 hello! 17:01:11 hello 17:01:11 Here we go again :) 17:01:19 apologies for the typo in the date yesterday 17:01:37 jwb: bahh, it happens 17:01:48 no biggie 17:02:05 sadly, i did it for 2 meetings 17:02:15 that's what i get for waiting until the last minute and rushing 17:02:43 * sgallagh notes that he sent out last week's agenda a few short hours before the meeting 17:03:06 is thozza on PTO? 17:03:33 hi t8m 17:03:42 hello all 17:03:56 ok, let's move on to the first topic 17:04:04 #topic #1178 Fedora 21 scheduling strategy 17:04:04 .fesco 1178 17:04:05 https://fedorahosted.org/fesco/ticket/1178 17:04:06 jwb: #1178 (Fedora 21 scheduling strategy) – FESCo - https://fedorahosted.org/fesco/ticket/1178 17:04:19 tbh, i forget why this is still in the meeting category 17:04:22 someone enlighten me? 17:05:15 i mean, afaik we only have tc6, it's broken, and we're still waiting for another TC to be spun 17:05:45 yeah, not much point in the ticket anymore I think. 17:05:52 Well, one thing 17:06:02 proposal: close unless there's some exceptional situation that comes up? 17:06:24 I've been saying this unofficially for a while, but maybe we should REALLY come out and say that Fedora isn't allowed to slip out of 2014? 17:06:43 sgallagh, and how would we realistically enforce that? 17:06:45 well, how do we do that? 17:06:47 As in, we'll start cutting out problematic pieces rather than slip again? 17:07:02 i'm not about to agree to that today 17:07:14 because we have no idea what "problematic pieces" actually means 17:07:28 I too would rather cut features than slip forever 17:07:38 guys... we aren't stalled on features 17:07:53 we're stalled on bugs 17:08:00 Not right now, but I trust that once the rel-eng stuff is resolved, it will stay that way. 17:08:06 like, arm netinstall doesn't work by the time of the release -- leave it out and do Alpha without this particular image 17:08:08 and you can't say "i'm going to cut out anaconda because it doesn't work on the workstation netinstall iso" 17:08:38 or workstation netinstall doesn't work -- ship alpha without it, just to be on time. 17:09:01 i don't disagree with the sentiment, but i don't think that's something we can actually say today with any kind of certainty. i'd rather address it as it comes up 17:09:14 "oh, we're blocked on arm netinstall? ok, vote on not blocking." 17:09:15 * nirik agrees with jwb. 17:09:19 I agree with Josh 17:09:25 we aren't down to just 1 thing either. 17:09:30 Sure, that's fine. 17:09:31 right 17:09:43 I was more just trying to set the stage for dealing with that if we get to it. 17:09:47 What is the summary of the current state of "rel-eng stuff"? 17:09:50 which, at the root of it, is really just nirik's proposal above 17:09:55 (Making sure we're willing to make those decisions) 17:10:09 jwb: Just slipping one more week is the very default path, though; the process more or less says "we slip" not "we ask FESCo whether to drop the feature" 17:10:21 https://qa.fedoraproject.org/blockerbugs/milestone/21/alpha/buglist 17:10:24 mattdm: I don't think Dennis has slept more than three consecutive hours since the original Alpha date :-/ 17:10:26 jwb, +1 17:10:43 mitr, well, FESCo can look at why we're suggesting to slip. so we can actually do what is proposed. we just have to do it. 17:11:21 we could also play with the time between milestones (which we did before) to shorten things... but I think that would need to be after alpha. 17:11:30 nirik, yes 17:11:52 * sgallagh nods 17:12:22 so for this specific ticket: 17:12:22 < nirik> proposal: close unless there's some exceptional situation that comes up? 17:12:37 +1 17:12:47 so, looking again over things... there is one thing that seems kinda large we could ask to just not do for alpha... 17:13:09 the per product netinstalls + mirrormanager changes for that. 17:13:16 +1 17:13:16 jwb/nirik: +1 17:13:34 aside from that I don't see anything we could really point to. 17:13:48 nirik, let's vote, then get back to that 17:13:56 +1 to the proposal 17:13:57 jwb: sure. Sorry to derail. 17:14:07 * nirik is +1 to his proposal for the record. 17:14:23 nirik: Umm, what's the alternative? 17:14:32 Just ship the ISOs and no net install trees? 17:14:48 sgallagh: right. say 'sorry, no netinstalls, we will have them for beta, thanks' 17:14:50 * sgallagh holds his question until after the vote 17:15:04 dgilmore, sgallagh ? 17:15:14 kalev, ? 17:15:38 +1 17:15:39 +1 17:16:10 * jwb waits 30 more secs 17:17:07 #agreed close unless there's some exceptional situation that comes up (+1:7, -1: 0, 0: 0) 17:17:16 #topic F21 Alpha status 17:17:33 ok, now you guys can talk about what we could possibly drop 17:17:34 :) 17:17:59 nirik, on your suggestion... it's somethign we could do but i fear that if we drop it for Alpha we'll not actually get it working for beta 17:18:14 well, I would only actually want to do that if it's the only blocker. 17:18:14 jwb: Why not? 17:18:27 if there's other things it seems not worth it. 17:19:11 sgallagh, because it seems like something that needs time to figure out, and it's important. if we decide it's not important for alpha, i'm concerned we'll worry about other more important things and it will eventually get pushed to the back burner 17:19:45 I suspect that this *specific* case won't have that problem 17:19:57 Since so many people rely on netinstalls 17:20:00 sgallagh, it also seems disjoint with the screams that workstation has to have a netinstaller because IMPORTANT, and then we drop it for alpha 17:20:02 We wouldn't let that slide 17:20:21 Yep, If something is not so important for alpha it does not mean it is not seriously important for beta and final 17:20:30 sgallagh, if it's a majorly important install method, then i think it needs to land in alpha so we have time to fix the bugs found in it before beta 17:20:54 my rather strong view is that _if_ we ship something, we need to make sure it actually works and not blame users for our bugs 17:21:23 so if we drop an image from being a Alpha release blcoker, then we have to also make sure to not ship that image to not give a bad public perception 17:21:37 i am working on the netinstall issues 17:21:48 we should be good by the end of teh week 17:21:59 dgilmore, to be clear, i don't think anyone is saying you aren't :) we're just discussing it's importance 17:22:06 its even 17:22:34 so since it's being worked on, and dgilmore says we should be ok, move on? 17:22:36 yeah, so, lets wait and see... if that ends up being the last/only blocker? 17:22:39 jwb: sure. it is important, I think we do want it for alpha 17:22:39 yeah. 17:22:42 +1 that. 17:22:44 cool 17:22:51 dgilmore: is there stuff we can get you help with? 17:23:13 (either directly related or to move other things that need to get done off your plate this week?) 17:23:22 mattdm: having people step up and help manage mirrormanager would be lovely 17:23:37 dgilmore: What does that entail? 17:23:56 dgilmore: I was wondering if I could ask puiterwijk about that... he dug into mm in the past, so he might be able to take a look. (I've not asked him tho) 17:24:10 nirik: can't hurt to ask 17:24:23 * nirik can do so 17:24:37 awesome 17:24:52 sgallagh: understanding how the code works to help do the tweaking to move links around 17:25:16 OK, then yes. Let's politely ask puiterwijk to bail us out :) 17:25:27 #action nirik to talk to puiterwijk about mirrormanager 17:25:57 ok, move on? 17:26:10 * nirik nods 17:26:12 #topic #1338 Non-responsive maintainer: masahase 17:26:12 .fesco 1338 17:26:12 https://fedorahosted.org/fesco/ticket/1338 17:26:13 jwb: #1338 (Non-responsive maintainer: masahase) – FESCo - https://fedorahosted.org/fesco/ticket/1338 17:26:59 ok, this seems fairly straighforward to me 17:27:01 +1 to orphaning packages so active folks can become point of contact. 17:27:10 +1 17:27:19 +1 17:27:26 OK, +1 17:27:49 +1 rubber stamp. looks like clear-cut example of process... proceeding. 17:27:54 just to be clear: is this orphaning both of his packages, or just the one that mooninite asks? 17:28:00 i would say both 17:28:16 me too, otherways the other one will just linger and not get any fixes 17:28:21 both 17:28:24 right 17:28:51 * nirik can do the orphaning and update the ticket... 17:29:02 thanks guys, thanks kevin :) 17:29:04 +1 17:29:59 sgallagh, dgilmore ? 17:30:16 I have another proposal after one is voted on. 17:30:27 Both. +1 17:30:40 mitr, ok. hold one sec 17:31:14 #agreed orphan masahase's packages so active people can become points of contact (+1: 7, -1: 0, 0: 0) 17:31:18 mitr, ok, go ahead 17:31:25 Proposal: In step 5 of http://fedoraproject.org/wiki/Policy_for_nonresponsive_package_maintainers , replace "to the Fedora devel list" with "to the FESCo trac". 17:31:38 #action nirik to orphan packages and update ticket 17:31:56 Rationale: it seems to happen from time to time that nobody from FESCo acts on that mailing list mail. (An alternative is that we just each individually promise to pay more attention to such mails.) 17:32:12 mitr, i'd suggest both actually 17:32:14 yeah, +1... I try and catch them, but sometimes I miss. ;( 17:32:19 mitr: +1 17:32:30 +1 this is just codifying the current practice 17:32:31 because the devel list mailing is for other packagers to get a heads up. the fesco ticket would be for us to pay attention 17:32:50 jwb: The public heads-up is step 4. 17:32:56 oh, hehe 17:32:58 yes, sorry 17:33:01 the procedure does already include a mail to devel in #4 --- this is a _second_ mail 17:33:06 ok, i'm good with your proposal then. +1 17:33:23 +1 17:33:28 (Current practice does tend to deviate from the formal policy, we tend to cut corners… but codifying that is for someone else to propose.) 17:33:55 +1 17:34:23 +1 17:34:43 #agreed In step 5 of http://fedoraproject.org/wiki/Policy_for_nonresponsive_package_maintainers replace "to the Fedora devel list" with "to the FESCo trac" (+1: 8, -1: 0, 0: 0) 17:34:53 mitr, care to update the wiki with that? 17:35:08 jwb: will do 17:35:16 #action mitr to update wiki 17:35:24 thanks. good proposal. 17:35:26 moving on 17:35:35 #topic #1339 missing acl's for some packages for epel7 17:35:35 .fesco 1339 17:35:36 https://fedorahosted.org/fesco/ticket/13389 17:35:36 jwb: #1339 (missing acl's for some packages for epel7) – FESCo - https://fedorahosted.org/fesco/ticket/1339 17:35:45 i'm fine with granting the ACLs for now 17:35:48 * nirik as noted in ticket was +1 17:35:51 im +1 to granting him access 17:35:52 +1 17:35:53 +1 17:36:20 +1 17:36:45 * nirik can do so if we approve 17:36:47 +1 17:37:12 +1 17:37:22 #agreed grant raveit65 epel7 ACLs for the relevant packages (+1: 8, -1: 0, 0: 0) 17:37:29 #action nirik to grant ACLs 17:37:48 ok, that's the end of the agenda 17:37:54 #topic Next week's chair 17:38:25 come on... 17:38:27 not everyone at once 17:38:42 * dgilmore should be able to 17:39:15 Nope. I'll do it so dennis doesn't have another thing. :) 17:39:18 dgilmore, you sure? your plate seems extra full lately 17:39:33 i'm using my chair privs to go with mattdm ;) 17:39:40 #info mattdm to chair next week 17:39:47 jwb: okay 17:39:49 thanks for offering though dgilmore 17:39:54 #topic Open Floor 17:40:03 who's got something else to chat about? 17:40:41 dgilmore: yes thanks. it's awesome that you are always willing to pick things up. 17:41:17 * jwb waits 2 min 17:42:55 ok, i guess we all have better things to work on :) 17:43:01 ending in 3 17:43:01 thanks jwb! 17:43:03 2 17:43:05 1 17:43:08 #endmeeting