17:02:24 #startmeeting F28 Final Go/No-Go meeting 17:02:24 Meeting started Thu Apr 26 17:02:24 2018 UTC. The chair is jkurik. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:02:24 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:02:24 The meeting name has been set to 'f28_final_go/no-go_meeting' 17:02:31 #meetingname F28-Final-Go-No-Go-meeting 17:02:31 The meeting name has been set to 'f28-final-go-no-go-meeting' 17:02:34 have fun ;) 17:02:37 .hello2 17:02:39 #topic Roll Call 17:02:39 lruzicka: lruzicka 'None' 17:02:40 .hello2 17:02:41 .hello2 17:02:41 .hello2 17:02:41 sgallagh: sgallagh 'Stephen Gallagher' 17:02:42 morning 17:02:44 jsmith: jsmith 'Jared Smith' 17:02:47 bowlofeggs: bowlofeggs 'Randy Barlow' 17:02:50 .hello mohanboddu 17:02:50 .hello jkurik 17:02:51 mboddu: mohanboddu 'Mohan Boddu' 17:02:53 * satellit_ listening 17:02:54 jkurik: jkurik 'Jan Kurik' 17:02:56 #chair nirik adamw sgallagh mboddu 17:02:56 Current chairs: adamw jkurik mboddu nirik sgallagh 17:03:00 .hello 17:03:00 kparal: (hello ) -- Alias for "hellomynameis $1". 17:03:05 .hello2 17:03:06 kparal: kparal 'Kamil Páral' 17:03:07 #topic Purpose of this meeting 17:03:14 #info Purpose of this meeting is to check whether or not F28 Final is ready for shipment, according to the release criteria. 17:03:19 #info This is determined in a few ways: 17:03:25 #info * No remaining blocker bugs 17:03:27 * mattdm is in right channel now. howdy everyone 17:03:30 #info * Release candidate compose is available 17:03:36 #info * Test matrices for Final are fully completed 17:03:46 * kparal pokes adamw 17:03:48 Hi everyone 17:03:53 #topic Current status 17:04:10 As far as I am aware, the RC for F28 Final is ready: http://dl.fedoraproject.org/pub/alt/stage/28_RC-1.1/ 17:04:11 .hello adamwill 17:04:12 adamw: adamwill 'Adam Williamson' 17:04:19 Test matrices are ready as well: https://fedoraproject.org/wiki/Test_Results:Fedora_28_RC_1.1_Summary 17:04:33 And we have some blockers to discuss: https://qa.fedoraproject.org/blockerbugs/milestone/28/final/buglist 17:04:42 Anyone wants to add something ? 17:05:11 looks like no 17:05:15 #info The RC for F28 Final is ready 17:05:21 #link http://dl.fedoraproject.org/pub/alt/stage/28_RC-1.1/ 17:05:26 #info Test matrices are available 17:05:32 #link https://fedoraproject.org/wiki/Test_Results:Fedora_28_RC_1.1_Summary 17:05:39 #info There are proposed and accepted blockers we need to go through 17:05:46 #link https://qa.fedoraproject.org/blockerbugs/milestone/28/final/buglist 17:05:58 Let's do Mini-blocker review 17:06:04 adamw: may I ask you please to chair the mini-blocker review ? 17:06:13 sure. 17:06:15 #topic Mini-Blocker Review 17:06:25 #link https://qa.fedoraproject.org/blockerbugs/milestone/28/final/buglist 17:06:58 #info there's one proposed blocker 17:06:59 #topic (1569211) GDM fails first login attempt after reboot 17:06:59 #link https://bugzilla.redhat.com/show_bug.cgi?id=1569211 17:06:59 #info Proposed Blocker, gnome-shell, NEW 17:08:01 hmm.. this looks like a serious one 17:08:01 so, we've been poking this one this morning. it seems fairly reproducible: shift (and possibly any modifier key?) isn't applied consistently to the first character entered in the gnome-shell password prompt, the first time it's used for any given password request. 17:08:09 it's...kinda weird 17:08:19 in that it seems that it always works fine the *second* time 17:08:32 or if you use backspace 17:08:36 so sgallagh is probably right that _mostly_ people are gonna assume they mistyped, do it again, and it'll work 17:08:37 right 17:08:46 Yeah. It looks like *general* thought is that we can live with this if we have a zero day update.... 17:08:49 * nirik doesn't think this is a blocker, but definitely a common bug + update fix. 17:08:53 are we likely to have that update? 17:09:06 mattdm: That discussion is still going on among the Workstation folks 17:09:07 i dunno, i'm kinda on the fence. on the one hand it's pretty bad, really. on the other, the practical impact of it is likely to be surprisingly not that big... 17:09:17 adamw: exactly 17:09:31 I've been seeing this consistently on Rawhide too. At least good to know I'm not *too* crazy. But yeah, it's survivable 17:09:42 mattdm: There's some question about it being an ibus issue or a gnome-shell one 17:09:45 puiterwijk: do you remember when it showed up on rawhide, by any chance? 17:09:45 Yeah, I thought it was my inability to type my password (on Rawhide)... 17:09:52 ... didn't know it was a real bug until just now :-) 17:10:10 adamw: I'm afraid not. I've thought I was ust crazy 17:10:11 I was glad of the excuse.. but I still type it wrong ;) 17:10:12 * jsmith is inclined to not make it a blocker, but to document it in Common Bugs 17:10:16 * satellit_ I use a Cap L first letter of password and have experienced this 17:10:17 I haven't experienced this ever while testing ... 17:10:21 jsmith, puiterwijk: That lends credence to the "most people will assume they just mistyped" line of thinking 17:10:29 sgallagh: yeah, agreed. 17:10:48 And if we get a fix at some point, it should be fine. Since installer images tend to not have passwords 17:10:50 * lruzicka has never used a capital letter first, so that might be the issue 17:10:51 lruzicka: it only happens when the first password character is capital (so you need to use shift) 17:10:52 sgallagh: While I agree with your sentiment, I don't think you can get away with thinking tha puiterwijk and I are in any way representative of "most people" 17:10:57 I like that we are all scoping our password entropy for the public :) 17:11:14 kparal: Oops. 17:11:20 my password begins with 🦄 17:11:21 langdon: I obviously just changed my password fater I told you this :) 17:11:21 jsmith: Yeah, you're people who are fare more likely to have discovered the truth on your own than the general public :) 17:11:22 a question was raised as to whether this would affect new passwords 17:11:29 jsmith: On the other hand, several people in the bug said the same thing :-) 17:11:34 because if it affects new passwords, that would be a blocker imo 17:11:41 bowlofeggs: it doesn't affect passwords set during install 17:11:53 what about passwords that are changed after install though? 17:11:58 i.e. I recently set a password with uppercase first letter in anaconda, and after the first login it works 17:12:01 What about passwords set using the "Users" section of the Gnome Control Panel? 17:12:07 I can try now 17:12:11 bowlofeggs: And we have confirmation that it doesn't affect the change-password dialog in a negative way. 17:12:17 puiterwijk: Please do :-) 17:12:20 If the bug prevented *ever* entering a password correctly this would be +1 blocker clearly. This seems like a minor annoyance and not a showstopper. 17:12:21 sgallagh: ah good 17:12:27 Or at least, only one of the two fields will have it happen, so they'll fail to match 17:12:33 So we won't be saving a bad password 17:13:21 Huh. It's also when unlocking things in gnome-control-panel 17:13:33 But the password I changed is recorded correctly 17:13:40 cool, i'm -1 then 17:13:53 So the password change is done correct. But gnome-control-panel has the same bug on first entry 17:14:00 So polkit has it too 17:14:19 Wow, I'm suddenly finding out I have not suddenly mistrained my fingers! 17:14:27 I'd like to have a reasonable confidence that we will be able to have a zero-day fix 17:14:28 I thought I'd been crazy for so long... 17:14:40 puiterwijk: you know these things aren't exclusive :) 17:14:46 .hello2 17:14:47 frantisekz: frantisekz 'František Zatloukal' 17:14:50 mattdm: oh yeah, I know 17:14:51 mattdm: Do we have the right people in the conversation to have that confidence of a zero-day fix? 17:15:13 What if not the first, but any other letter is capital? Does it also behave that way? I cannot try right now. 17:15:16 Do we have anyone from gnome here? 17:15:18 jsmith: The right people are aware of the BZ, but I'm not sure they're present in this meeting 17:15:19 mattdm: As far as I can tell, we haven't even yet identified which component could be causing the issue. 17:15:33 it's too early to tell much here I fear. 17:15:59 i kinda suspect it'll get figured out, but hard to tell for sure. 17:16:01 lruzicka: it's just first character 17:16:40 puiterwijk: interesting 17:16:45 While this bug is clearly annoying, I don't think it's enough on its own to stop the release. So -1 from me. 17:16:55 * nirik is -1 on blocker also 17:17:00 -1 17:17:00 -1 from me as well 17:17:10 * satellit_ -1 17:17:19 I believe that it only affects passwords with the first capitalized letter, it will not be that many. -1 from me. 17:17:23 -1 blocker. +1 happy I've found this, and +50 happy I'm not mistyping my password twenty times a day! 17:17:33 -1 17:17:34 Its annoying but we can live with it, -1 Blocker 17:17:38 lruzicka: capitalized and anything that needs shift 17:17:56 i.e. it looks like shift is ignored in the first character of a password entry the first time 17:17:59 Okay, my "ship it" bias is leaning me into the consensus 17:18:02 puiterwijk: but only at the first position, right? 17:18:06 lruzicka: correct 17:18:08 mattdm++ me too. 17:18:10 I was just typing that out :) 17:18:25 * puiterwijk now changes his password away from the tes tvalues to prevent forgetting it.... 17:18:31 i'm actually curious if it affects any other necessary modifiers too (alt-gr modifies the character typed in some keyboard layouts for e.g., i think) but eh 17:19:08 adamw: I am using alt gr in passwords, but never at the first position, so I cannot tell here 17:19:17 adamw: Is "eh" a "-1 blocker" vote? 17:19:20 I can probably try that if people want to know 17:19:32 I'm now forgetting my password anyway... One more attempt won't make it worse 17:19:41 * sgallagh notes that the GNOME folks in the BZ also voted -1 blocker 17:19:52 proposed #agreed 1569211 - RejectedBlocker (Final) - there's a general feeling that this is bad but not *so* bad that it constitutes a serious enough criteria violation to abort the release at this point. most users who run into this will just assume they mistyped, and type again; it's not likely to lock anyone out of anything. We certainly hope that a fix for this can be made available as a 0-day blocker 17:19:54 mattdm: As for the "get it fixed soon" part -- I can guarantee you I will be among the people seeking a quick fix here. And I intend to be transparent about the problem in advance, to show the exact tiny corner we cut to make GA. (To be clear, I support -1 blocker 100% and would say that too.) 17:20:09 adamw: ack 17:20:32 I have utmost confidence in mclasen & crew to get this sussed out and a fix ready 17:20:32 adamw: ack 17:20:33 ack 17:20:34 ack 17:20:35 ack 17:20:38 stickster++ 17:20:38 sgallagh: Karma for pfrields changed to 19 (for the f27 release cycle): https://badges.fedoraproject.org/tags/cookie/any 17:20:39 ack 17:20:44 puiterwijk just write it down 17:20:46 ACK 17:20:47 ack 17:20:47 ack 17:20:55 .fire langdon 17:20:55 adamw fires langdon 17:20:56 * mclasen looks up 17:20:57 stickster_temp: yes, works for me. 17:21:05 langdon: no. 17:21:34 ack 17:21:34 #agreed 1569211 - RejectedBlocker (Final) - there's a general feeling that this is bad but not *so* bad that it constitutes a serious enough criteria violation to abort the release at this point. most users who run into this will just assume they mistyped, and type again; it's not likely to lock anyone out of anything. We certainly hope that a fix for this can be made available as a 0-day blocker 17:21:36 langdon: How does Russ Doty say it? "A secure environment is one where the password sticky-note is in a drawer and in an ultra-secure environment the drawer is also locked" 17:21:45 * puiterwijk would find the inability to unlock a resumed system worse. But I have only hit that wit hrawhide 17:21:48 mclasen: q.v. the password bug we are not blocking GA on ... looking for a zero-day fix but I'm sure you know it already 17:21:54 (not the fix, that people are looking) :-D 17:21:55 sgallagh: lol.. but the key should be in the lock of course 17:22:00 puiterwijk: it affects unlock, we confirmed that in the bug. 17:22:03 langdon: I might actually have a sticky note on my display with a password. That passowrd might or might not actually work :) 17:22:08 of course, second time you type it works. 17:22:13 adamw: right. I just mean that after resume, I cannot unlock *at all* 17:22:21 the shield just doesn't go away. But that's entirely different 17:22:23 good explanation adamw. 17:22:33 puiterwijk: lol 17:22:53 stickster_temp: I saw something in passing, but I haven't looked myself 17:23:03 puiterwijk: It probably works *somewhere*. There are billions of humans and we're not actually all that imaginative 17:23:23 puiterwijk: oh, yeah. that's something else. 17:23:24 you can always tell the correct sticky note because it is at the bottom of the pile of sticky notes 17:23:33 sgallagh: hah. I think I have a photo of the post-it somewhere. You can try to find where it works :) 17:23:45 OK, so shall we move on? 17:23:53 adamw: are we going to discuss "#1537253 Add-On Modularity" ? 17:23:58 mclasen: I think there is a question as to what subsystem the bug exists in (ibus? gnome-shell?) -- but needs investigation -- BZ 1569211 17:23:59 sure. 17:24:02 17:24:03 adamw: yeah. And as said, I've only tried rawhide. Just never got to trying it on f28 yet. Just really, really annoying. (and I'll file it as proposed blocker for f29) 17:24:05 #info we have one outstanding accepted blocker 17:24:16 ibus has no role in a password entry 17:24:20 puiterwijk: i've been suspending and resuming f28 just fine, but if you want to test it... 17:24:25 at least in gtk 17:24:36 mclasen: this is gnome-shell's password entry, not gtk's. 17:24:37 adamw: I don't have f28 here, and I'm !home at the moment. But it's basically after undocking 17:24:42 mclasen: we're discussing it in #fedora-desktop 17:24:51 mclasen: the ibus possibility was suggested by garnacho 17:24:58 puiterwijk: oh, docking fun. 17:24:59 So I lock, undock, open lid, and the shield just won't go away 17:25:04 adamw: yep! 17:25:12 Anyway, as said, unrelated to password entry. 17:25:14 -1 to docking problems. focccusssssss :) 17:25:16 * mclasen moves on and lets the experts handle it 17:25:31 #topic (1537253) Add-On Modularity 17:25:31 #link https://bugzilla.redhat.com/show_bug.cgi?id=1537253 17:25:31 #info Accepted Blocker, Changes Tracking, VERIFIED 17:25:44 so, the modularity seems to be working as expected now 17:25:57 yep! 17:26:03 -1 to block on this 17:26:15 or rather: blocker resolved! 17:26:15 #info once again for the record, FESCo has declared the Add-on Modularity feature as 'blocking' F28 release, this bug is being used to track whether the feature meets FESCo's stated requirements 17:26:35 mattdm: yeah.. thats what i was confused about 17:27:25 sgallagh: should we just make is CLOSED? 17:27:26 right. Can close if it's been verified? 17:27:27 there was a test day for this one, the results seemed pretty promising, as I remember 17:27:35 sumantro: What do you think? 17:27:47 -1 to block on this one 17:28:01 With both me and sumantro reporting it as passing the criteria, I'd say this is CLOSED CURRENTRELEASE 17:28:15 sgallagh: ack 17:28:18 sgallagh: I agree 17:28:21 ack 17:28:24 sgallagh: cool 17:28:30 * sumantro nods in agreement 17:28:31 We're not voting on whether it's a blocker. It is, FESCo said so. We're accepting it as fixed :) 17:28:44 right 17:28:44 sgallagh: Yes, I agree with that. 17:29:10 +1 to close currentrelease 17:29:37 #info multiple testers report that Modularity does meet FESCo's requirements, so this blocker is considered addressed. 17:29:56 adamw++ 17:29:58 langdon: Karma for adamwill changed to 21 (for the f27 release cycle): https://badges.fedoraproject.org/tags/cookie/any 17:30:04 we don't really need to worry about closing it or whatever (we'd have to check the Change process for when Change bugs are supposed to be closed), it just needs to be on record that it's addressed. 17:30:48 adamw: shall I move on ? That is all from the blockers, right ? 17:30:50 adamw: I closed it. It seems OK to do that as with any other bug. 17:31:05 that's all the blockers, yes. 17:31:13 #topic Test Matrices coverage 17:31:15 #link https://fedoraproject.org/wiki/Test_Results:Fedora_28_RC_1.1_Summary 17:31:15 Yay! 17:31:17 yaay! 17:31:29 so, how do we see the testing coverage ? 17:31:51 zbyszek: well, https://fedoraproject.org/wiki/Changes/Policy should really document how and when Change bugs are closed, but it apparently...doesn't. 17:32:01 it looks pretty good. 17:32:13 jkurik: Usually with our eyes, or our fingers if using a braille device. 17:32:46 ha. someone did the sas test. Kudos. 17:32:51 note, i'm in the middle of re-doing the desktop_menus test for KDE, as i don't think "I haven't tried all applications" is acceptable (the whole point of the test is to try all the applications) 17:33:11 * satellit_ note notes on my testing https://fedoraproject.org/wiki/Test_Results:Fedora_28_RC_1.1_Summary#Default_boot_and_install_.28x86_64.29 17:33:11 FCoE is missing... thats a pretty fringe one tho 17:33:12 * adamw finds a working NNTP server. the things i do for my sins. 17:33:50 nirik: we do have a story for FCoE...it's https://bugzilla.redhat.com/show_bug.cgi?id=1557655 17:34:16 QA:Testcase_realmd_join_kickstart missing... hopefully that one is ok 17:34:20 https://bugzilla.redhat.com/show_bug.cgi?id=1557655#c28 tends to indicate it actually worked at 2018-04-10, though i should've asked lili to be a bit more explicit. 17:34:27 nirik: hum, coconut should test that. 17:34:29 * adamw checks' 17:34:39 AD one also 17:34:48 QA:Testcase_domain_client_authenticate (Active Directory) 17:34:50 adamw: i think microsoft has one :) 17:35:11 nirik: it worked, https://openqa.fedoraproject.org/tests/229679 . i'll have to figure out why the result didn't make it to the wiki. 17:35:29 nirik: sgallagh usually does the AD tests 17:35:30 sgallagh? 17:35:31 adamw: maybe token expiry/blanking before you noticed? 17:35:42 nirik: I can probably run the kickstart one in a few minutes if we're worried about it. 17:35:43 you could probably get it from one of the warez news sites 17:35:43 puiterwijk: i ran a blanket result report after we fixed that 17:35:49 langdon: oh, nntp server :P 17:35:52 adamw: ah, okay. Then I don't know :) 17:36:40 nirik: But in general if the cockpit one works and kickstart isn't otherwise broken, it's going to be fine 17:36:56 sgallagh: how about the AD test? 17:37:05 nirik: I did all of the other AD tests 17:37:09 Did they not get marked? 17:37:50 langdon: for the record, news.mixmin.net appears to work. :P 17:38:12 sgallagh: we have had the kickstart test fail where others didn't in the past for...reasons. the installer env can strip files it needs to work, etc. 17:38:14 Ah, looks like I forgot to mark QA:Testcase_domain_client_authenticate (Active Directory) but it passed 17:38:17 adamw: yeah.. i think i mirror that one ... 17:38:25 * langdon kids 17:38:27 sgallagh: yeah, that one. cool. 17:38:37 nirik: That one is also Optional, FWIW 17:38:55 adamw: OK, do we want to hold up the meeting while I run that test? 17:38:56 wow, someone posted to alt.os.linux.fedora on 2017-12-26 17:39:01 Should take about 10 minutes 17:39:12 adamw: did mattdm reply? 17:39:13 oh, yeah, anything Optional is ignorable. 17:39:19 langdon: heh 17:39:28 adamw: did they upload pirated copies of our isos? 17:39:38 * langdon wonders if matrix bridges nntp 17:39:39 puiterwijk: no, actual sane message about apper crashing. 17:39:39 (I've been told that that's the only use of NNTP these days) 17:39:47 you were lied to! 17:39:55 Wow! 17:40:12 sgallagh: you test, i'll keep them distracted with inane chatter 17:40:13 puiterwijk: specifically pirated open source software?!?! 17:40:20 adamw: i got you 17:40:25 nirik: how about we setup usenet.fedoraproject.org, and move all mailing lists there? 17:40:30 langdon: of course! 17:40:34 puiterwijk++ 17:40:45 puiterwijk++ 17:40:51 puiterwijk, but, are you going to write basset for nntp? 17:40:53 sure. we can use uucp to transfer the posts to other nodes. 17:41:04 puiterwijk says to adamw: "here, hold my beer" 17:41:14 Are we still meeting or is this now just fun chat? I'm OK either way, just want to know whether I should wander off. 17:41:22 stickster: we're waiting for some tests I think. 17:41:27 puiterwijk: shall not we run our own BBS ? 17:41:33 stickster: fun chat until sgallagh comes back with his test results 17:41:48 okey doke, back in a few min 17:41:50 if we run a BBS we need to host Exitilus (spelling?) 17:41:51 or LORD 17:41:52 jkurik: I have some (bad) memories of phpBB. 17:41:55 jkurik: single phone line please.. 17:42:02 stickster: i'm stalling while i run the kde menus test and sgallagh runs a kickstart test. 17:42:27 * puiterwijk does note that people probably want to be careful with encouraging me to do things... We might actually have usenet.fp.o before the end of the day... You can ask about anyone that I'll do it. 17:42:28 oh good, KDE is up to shipping *three* web browsers on the live image now. this is certainly sane. 17:42:29 well, openqa did pass the kickstart test I thought? 17:42:39 nirik: openqa tests freeipa, not AD 17:42:56 testing AD in openqa is kinda tricky because windows 17:43:04 but sgallagh already did those? 17:43:07 adamw: oh, so workstation is lacking? I guess I'll need to find more browsers to add to it then. I'm sure we can go over 3 17:43:11 * nirik is confused which test sgallagh is doing 17:43:12 he didn't do the kickstart one apparently. 17:43:17 nirik: kickstart one 17:43:19 I'm doing the kickstart test right now 17:43:20 * stickster thinks back fondly to PCBoard 17:43:21 nirik: kickstart enrolment to an AD server. 17:43:22 Please hold 17:43:25 puiterwijk: sudo dnf emerge usenet 17:43:27 -y 17:43:34 Oh, and WWIV. I liked that one a lot. 17:43:34 bowlofeggs: ... yeah. that. 17:43:40 [10:38:14] Ah, looks like I forgot to mark QA:Testcase_domain_client_authenticate (Active Directory) but it passed 17:43:50 nirik: we've tested enrolment to an AD server in ways other than kickstart, and we've tested enrolment to a FreeIPA server via kickstart. but we haven't tested AD enrolment via kickstart. :P 17:43:53 For those who want something scary: https://github.com/puiterwijk/dnf-plugins-emerge 17:43:58 nirik: that's not the kickstart test. 17:44:01 nirik: That's a different one I forgot to mark 17:44:05 puiterwijk: DO. NOT. WANT. 17:44:11 ok. 17:44:30 stickster: I warned people: be careful with jokes around me. It actually works :) 17:45:19 has anyone written a BBS (or nntp server for that matter) in js yet? 17:45:27 puiterwijk: count me in as a volunteer for testing 17:45:29 lol 17:45:34 (at JS bbs) 17:45:56 jkurik: Hah. You can file issues :) 17:45:57 langdon: more like, has anyone written and rewritten 10 frameworks in JS for BBS's? 17:46:40 we have a linux server in js.... kernel through bash.. i think it is a serious oversight that there is not a BBS or NNTP server for it 17:46:55 puiterwijk: That is something that I think I will never use, but probably test it for fun :) 17:47:11 langdon: something like https://nodebb.org/ ? 17:47:35 puiterwijk: NICE 17:47:36 mboddu: yeah... Please don't use it on any serious system... It is probably somewhat reasonably sane, but you can get a seriously confused system 17:47:54 * puiterwijk gets memories of phpBB and closes that page and part of his memory 17:47:58 ha 17:48:22 puiterwijk: Lets test f28 along with that and start filing bugs for adamw :D 17:48:23 OMG.. it does not appear to be open sourcE?!?! 17:48:29 mboddu: haha. 17:48:38 kde also ships a viewer for an ancient proprietary outlook attachment format because of course it does 17:48:53 langdon: https://github.com/NodeBB/NodeBB 17:49:15 phew.. 17:49:17 langdon: puiterwijk: you want this, https://www.wwivbbs.org/ 17:49:31 stickster: I am pretty sure I do not 17:49:35 lol 17:49:41 Test is almost done, looking good so far. 17:49:45 I'm sure you want it just as much as nodeBB 17:49:47 * langdon goes to deploy 17:49:59 * puiterwijk waits for "dnf module enable nodebb" 17:50:05 or "dnf module enable emerge" 17:50:14 langdon: ^ there's your cue 17:50:18 dnf module enable yak 17:50:19 dnf emerge nodebb? 17:50:45 oh, heads up that nirik and I have a meeting in 10 min 17:50:58 * nirik nods 17:51:02 is it about deploying a bbs? cause if not, you should just cancel 17:51:07 * adamw checks how many more apps there are in this interminable goddamn list 17:51:17 langdon: haha :) 17:51:24 if only 17:51:35 adamw needs a bot minion 17:51:52 stickster: i have several, they're just not very good at this. 17:51:59 need moar minions 17:52:04 seriously, who came up with this open source is a good thing idea .... now we have all this software!! 17:52:13 .fire langdon 17:52:13 adamw fires langdon 17:52:19 nirik: if you go for a meeting, you might propose your Go or No-Go in advance :-) 17:52:20 huh, I wonder if we could get somethiing like sellinium in jenkins to do these test 17:52:24 langdon: it's for choice 17:52:27 http://www.islinuxaboutchoice.com/ 17:52:37 * langdon notes he has been fired twice in this meeting.. hmmm 17:52:43 * sgallagh wonders if he can .fire his boss... 17:53:01 sgallagh: you can try, it's all about transparency 17:53:05 haha 17:53:11 360 reviews! 17:53:18 nirik: this is fundamentally a difficult test to automate because there are hundreds of apps on the workstation and kde lives, the loadout can and does change without warning, and *any* app can change its behaviour 17:53:18 sgallagh: I've had the reverse happen to me on IRC. And then got an explanation that it wasn't serious, just to make sure :) 17:53:24 also just testing 'basic functionality' of some apps is tricky 17:53:35 Should we consider dropping the "test all the applications in the menu" criteria for kde in specific, given the large list of apps and the trickiness of testing? 17:53:38 jkurik: sure. I see no blockers... if my capacity as fesco person, releng person, infra person I am all go. :) 17:53:38 yeah, for sure 17:53:46 OK, kickstart test has passed 17:53:49 *phew* 17:53:49 sorry, that was in response to adamw, not mattdm 17:53:51 why are most of the best rants preserved on the internet by ajax? 17:53:53 nirik: basically an 'automated' test which has to run and do some basic thing in 100+ apps is going to break *all the damn time* no matter whether its done in openqa or selenium or dogtail or anything 17:53:56 cool, thanks sgallagh 17:54:08 nirik: thanks 17:54:10 adamw: true. 17:54:10 was that the only one we were waiting on? 17:54:21 mattdm: it's a possibility. i would rather we try and get the kde folks to cut down the number of freaking things they ship 17:54:29 stickster: i'm still mining through kde menus 17:54:32 oh yeah 17:54:37 that thing you keep saying sucks 17:54:38 i'm about 60% done, haven't come across anything catastrophic yet 17:55:01 we can probably just go ahead and sign the damn thing off 17:55:12 Sounds good to me! 17:55:13 * mboddu cant keep up with the chat here 17:55:18 i doubt anyone's going to block on "657th obscure app shipped by kde crashes" at this point anyway 17:55:18 * Southern_Gentlem thinks we need to have the SIGs do this testing in the future 17:55:24 Southern_Gentlem: i've been trying that for years. 17:55:29 adamw: are you just smoking each? or sometihng more? 17:55:31 adamw: Yeah, because if you come up with anything, it's going to have to be epicly bad for me not to say "SHIP IT ANYWAY" 17:55:45