16:00:37 #startmeeting f20alpha-blocker-review-4 16:00:37 Meeting started Mon Sep 9 16:00:37 2013 UTC. The chair is tflink. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:37 Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:00:37 #meetingname f20alpha-blocker-review-4 16:00:37 The meeting name has been set to 'f20alpha-blocker-review-4' 16:00:37 #topic Roll Call 16:00:52 who's ready for some blocker review fun? 16:01:02 #chair kparal pschindl 16:01:02 Current chairs: kparal pschindl tflink 16:01:09 * kparal here 16:01:12 * pwhalen is here 16:01:23 *me* 16:01:24 * pschindl is here 16:01:33 * nirik is lurking, ping if I can help with anything 16:01:56 lots of people today :) 16:02:47 * jreznik is also here :) even more people today! 16:02:50 * mkrizek is here 16:02:56 * satellit_e listening 16:03:32 time for some boilerplate 16:03:35 * roshi lurks 16:03:36 #topic Introduction 16:03:40 Why are we here? 16:03:41 #info Our purpose in this meeting is to review proposed blocker and nice-to-have bugs and decide whether to accept them, and to monitor the progress of fixing existing accepted blocker and freeze exception bugs. 16:03:46 #info We'll be following the process outlined at: 16:03:46 #link https://fedoraproject.org/wiki/QA:SOP_Blocker_Bug_Meeting 16:03:51 #info The bugs up for review today are available at: 16:03:51 #link http://qa.fedoraproject.org/blockerbugs/current 16:03:57 #info The criteria for release blocking bugs can be found at: 16:03:57 #link https://fedoraproject.org/wiki/Fedora_20_Alpha_Release_Criteria 16:04:02 #info Up for review today, we have: 16:04:11 #info 5 Proposed Blockers 16:04:12 #info 7 Accepted Blockers 16:04:12 #info 1 Proposed Freeze Exceptions 16:04:12 #info 6 Accepted Freeze Exceptions 16:04:29 if there are no objections, we'll start with the proposed blockers 16:05:04 #topic (1005249) netinst with text mode doesn't allow to install GNOME 16:05:07 #link https://bugzilla.redhat.com/show_bug.cgi?id=1005249 16:05:10 #info Proposed Blocker, anaconda, NEW 16:05:38 * kparal assumes he's the secretary today 16:05:54 oh yeah, I forgot to ask for a volunteer 16:05:55 thanks 16:06:57 kparal: could this have been a mirror issue? which dvd did you use? 16:07:05 mirror/package issue 16:07:38 TC4 16:07:46 but there haven't been any errors which would have caused that in TC4 16:07:47 reproduced several times in a row 16:08:07 I don't think this is a mirror issue 16:09:10 * kparal reproducing again 16:09:17 do it ;) 16:09:44 Viking-Ice: ? 16:09:49 damn, it works now with TC5 :) 16:09:52 anyone else able to reproduce it (maybe hitting different mirror etc.)? 16:09:59 tflink, kparal test it 16:10:05 mean testing it 16:11:02 * kparal trying TC4 netinst 16:11:03 so punt for more data or close ? 16:11:20 as described, I'm +1 16:11:31 if it works in TC5 - more close? would be better to have more people trying it 16:11:42 punt it 16:11:44 * jreznik would try, just my kvm stopped working for some reason 16:11:48 it works for me with TC4 at the moment 16:12:10 jreznik, modprobe kvm kvm_intel perhaps ;) 16:12:11 so I'll re-try some more and we can discuss on wednesday 16:12:20 heisenbug! 16:12:32 sounds that mirror/package issue 16:12:37 like that 16:12:43 yep 16:12:52 proposed #agreed 1005249 - AcceptedBlocker - Violates the following F20 alpha release criterion for a text based install: " 16:12:55 When doing a graphical install using the dedicated installer images, the installer must be able to install each of the release blocking desktops, as well as the minimal package set.: 16:12:59 damnation 16:13:14 nack let's punt this for more data points 16:13:19 nack 16:13:29 proposed #agreed 1005249 - AcceptedBlocker - Violates the following F20 alpha release criterion for a text based install: "When doing a graphical install using the dedicated installer images, the installer must be able to install each of the release blocking desktops, as well as the minimal package set." 16:13:33 ok 16:13:39 I thought punt? 16:13:51 punt and ask more people to retest 16:14:00 * kparal brb in 2 min 16:14:10 kparal, pls, comment it in bz too 16:14:35 proposed #agreed This may end up being a package/mirror issue and needs more data. will revisit at the next review meeting 16:14:47 ack 16:15:05 ack 16:15:07 ack 16:15:09 ack 16:15:09 ack 16:15:14 #agreed This may end up being a package/mirror issue and needs more data. will revisit at the next review meeting 16:15:22 #topic (1005251) Prepare Fedora 20 wallpaper 16:15:22 #link https://bugzilla.redhat.com/show_bug.cgi?id=1005251 16:15:23 #info Proposed Blocker, distribution, NEW 16:16:00 +1 16:16:10 proposed #agreed 1005251 - AcceptedBlocker - Violates the following F20 alpha release criterion: "The default desktop background must be different from that of the two previous stable releases." 16:16:45 ack 16:16:47 ack 16:16:49 ack 16:16:53 ack 16:16:54 ack 16:17:12 we have design, martin will work later today on packaging, has to be coordinated with kde theme 16:18:12 #agreed 1005251 - AcceptedBlocker - Violates the following F20 alpha release criterion: "The default desktop background must be different from that of the two previous stable releases." 16:18:20 #topic (1005232) Update Fedora 20 release name to Heisenbug 16:18:20 #link https://bugzilla.redhat.com/show_bug.cgi?id=1005232 16:18:20 #info Proposed Blocker, fedora-release, ON_QA 16:18:59 tried it. It's set on updated system. 16:19:17 proposed #agreed 1005232 - AcceptedBlocker - Violates the following F20 alpha release criterion: "Any component which prominently identifies a Fedora release version number, code name or milestone (Alpha, Beta, Final) must do so correctly." 16:19:48 ack 16:19:57 ack 16:19:58 ack 16:20:05 ack 16:20:14 #agreed 1005232 - AcceptedBlocker - Violates the following F20 alpha release criterion: "Any component which prominently identifies a Fedora release version number, code name or milestone (Alpha, Beta, Final) must do so correctly." 16:20:19 ack 16:20:21 #topic (1002464) cannot unlock gnome lock screen 16:20:21 #link https://bugzilla.redhat.com/show_bug.cgi?id=1002464 16:20:21 #info Proposed Blocker, gnome-shell, MODIFIED 16:20:48 -1 blocker +1 fe 16:21:03 why -1 blocker? 16:21:08 I don't think this is blocker. There is easy work around. 16:21:19 but unobvious 16:21:23 kparal: criterion? 16:21:28 but switching to second console and back isn't user friendly 16:21:35 that's true. But we don't have criterion for this 16:21:50 definitely +1 FE 16:21:53 yes. It is ugly, that's why I'm +1 FE 16:22:00 this is ugly and does affect lives 16:22:01 -1 blocker +1 FE 16:22:08 we don't have criteria for every piece of the stack 16:22:17 and this is alpha people should expect bugs 16:22:26 tflink: does it really affect Lives? it works for me on Live 16:22:29 because there's no password 16:22:31 that systemd build changes quite a bit, though 16:23:21 ? 16:23:22 kparal: c#4 claims that it affects lives 16:23:49 that's gnome shell problem not systemd 16:23:59 at least this bug never landed on our table 16:24:04 * kparal booting Live 16:24:09 Viking-Ice: the fix is in systemd 16:24:31 tflink: Live is working ok, since there's no password prompt 16:24:39 with locked screen 16:24:47 then I'm not even sure about FE 16:24:57 tflink: unless you create your own user, of course 16:25:11 Viking-Ice: https://bugzilla.redhat.com/show_bug.cgi?id=1002464#c5 16:25:12 I think it's a violation of " A system installed with a release-blocking desktop must boot to a log in screen where it is possible to log in to a working desktop using a user account created during installation or a 'first boot' utility. " 16:25:16 hmm might have been this logind: update state file after generating the session fifo, not before 16:25:28 kparal, not with gnome lock screen 16:25:29 kparal: I disagree, you can login just fine and apply updates 16:25:34 I can't log in twice, it's a different variation of the criterion 16:25:46 login twice or unlock screen after logging in? 16:25:49 kparal, dont lock your screen then ( workaround 16:25:51 unlock 16:26:20 I don't like releasing Alpha that breaks for _everyone_ in the first 5 minutes 16:26:30 of inactivity 16:26:32 I'm not thrilled by it either 16:26:41 yes, sure, some people will search for workaround 16:26:45 many won't 16:26:53 but it's not like we'd be taking a systemd build which only fixes this bug 16:27:01 yeah 16:27:04 and this is alpha 16:27:13 we don't even have RC yet 16:27:26 desktop experience not being pristine is not high priority at this point 16:27:36 oof, that hadn't occurred to me yet 16:27:42 the RC part 16:27:47 I'd definitely take the fix, even if not accepted as a blocker 16:27:51 we really need that anaconda build today 16:27:57 but I digress 16:28:34 for blocker, I see +1, -3 16:28:50 for FE, I see +2 16:28:56 +1 FE from me 16:29:04 oh, +3 16:29:05 obviously :) 16:29:11 other votes? 16:29:20 +1 FE 16:29:22 +1 FE 16:29:23 * tflink is -1/+.5 16:29:29 +1 FE 16:29:37 so let's propose -1/+1 16:29:38 can we get some people testing that systemd build? 16:30:11 I'll make sure it's tested 16:30:12 * jreznik is more -1/0 - systemd is systemd and as Viking-Ice said - Alpha is not much about desktops... 16:30:26 but if kparal makes sure it gets some coverage... 16:30:35 * jreznik trusts him :) 16:30:58 proposed #agreed 1002464 - RejectedBlocker AcceptedFreezeException - While this doesn't violate any of the F20 alpha release criterion, it is a rather ugly problem that's easy to hit. A tested fix would be considered past freeze 16:31:05 ack 16:31:07 ack 16:31:07 ack 16:31:08 ack 16:31:10 ack 16:31:24 #agreed 1002464 - RejectedBlocker AcceptedFreezeException - While this doesn't violate any of the F20 alpha release criterion, it is a rather ugly problem that's easy to hit. A tested fix would be considered past freeze 16:31:46 #topic (1004902) Error in `/usr/bin/sddm': double free or corruption (!prev): 0xb7b06c28 16:31:49 #link https://bugzilla.redhat.com/show_bug.cgi?id=1004902 16:31:52 #info Proposed Blocker, sddm, MODIFIED 16:32:15 that's an blocker 16:32:19 I didnt include which criteria "A system installed with a release-blocking desktop must boot to a log in screen where it is possible to log in to a working desktop using a user account created during installation or a 'first boot' utility. " 16:32:19 hence +1 16:32:22 it's pretty easy to workaround 16:33:02 Martin changes the order for now in the update, so KDE Plasma Workspaces is first, it would crash on custom thought (but not default option anymore) 16:33:32 workaround or not it's an clear violation of the critera 16:34:02 not saying, I'm not probably +1 16:35:42 what is the workaround? 16:35:56 what "Custom" means? 16:36:07 tflink: select "KDE Plasma Workaround" instead of preselected Custom 16:36:13 kparal: no idea 16:36:20 great :) 16:36:36 it's hidden now in the updated package 16:36:54 ok, I see, the login screen had pre-selected invalid value. right? 16:36:56 +1 blocker. There is unknown option :) 16:37:01 now it has pre-selected correct value, right? 16:37:23 kparal: after update, it should be correct value 16:37:47 sounds as a blocker, since KDE on ARM is primary desktop, IIRC 16:37:54 jreznik: and Custom will be still present? 16:38:01 apparently, it affects x86_64 as well 16:38:03 kparal: it's not only about ARM 16:38:05 kparal, not just arm 16:38:09 garretraziel: no 16:38:21 the more reason to take it as a blocker then 16:38:45 I'm +1 fe, not as sure about blocker since there is a pretty easy workaround 16:38:47 +1 16:39:05 jreznik: was Custom pre-selected even for x86 installs, or just available? 16:39:25 the workaround is pretty clear to figure out 16:39:48 kparal: it was 16:39:59 (preselected, as far as I understand it) 16:40:23 mkrizek: so screen just flashes and you get the idea to change the option? or does it crash hard? 16:40:59 kparal: I had to reboot IIRC 16:41:22 pwhalen: can you confirm? 16:41:23 kparal, just booted me back out to the log in screen, didnt crash but it wasnt clear I needed to change something 16:41:25 * jreznik is becoming more +1 blocker, it's easy to workaround but you have to know about workaround 16:41:25 The criterion talks about possibility of log in and there is possibility. So I'm more +1 for FE 16:41:56 pschindl: there's always possibility of fixing the source code and recompiling yourself ;) 16:41:57 it's a grey area 16:42:01 does it affect lives? 16:42:20 tflink: there should be autologin on lives 16:42:25 this is not a gray are it's an clear violation for an release blocking desktop 16:42:28 * kparal trying KDE Live 16:42:40 Viking-Ice: but you _can_ login 16:42:48 kparal: But you need just two clicks to change session 16:42:58 why are we even discuss this just close this as a blocker and let's pull in the fix 16:43:04 with such a simple fix, I think we're spending too much time discussing 16:43:07 KDE Live uses autologin 16:43:21 yeah, I just wasn't sure if autologin was affected 16:43:25 however, if I log out from Live... 16:43:44 is anyone -1 blocker? 16:43:54 if I log out, I'm back in the session immediately :) 16:43:58 jreznik, I think was the only one 16:43:59 I'm +0.5/+1 16:44:06 " I'm not probably +1" 16:44:30 Viking-Ice: no, it wasn't -1 16:44:38 -1 and not+1 aren't the same 16:44:58 unless Mr. Bool says so 16:44:58 tflink: not saying, I'm not probably +1 16:45:10 that's what I said :) 16:45:15 in the world of black and white it is and we also just count whole numbers so I dont se point in people voting 0.5 or 0.47 16:45:16 *Boole 16:45:17 I am +1, because it clearly violates the criteria 16:45:17 with context it means 16:45:45 I don't see anyone arguing against blocker, so 16:47:18 Ok. There is fix. It violates the criterion. +1 for blocker. 16:47:20 ok, so are we going with blocker? let's move on 16:47:31 proposed #agreed 1004902 - AcceptedBlocker - While not a 100% clear-cut violation of the criteria for KDE, this was deemed close enough to a violation to take as a release blocking bug for F20 alpha: "A system installed with a release-blocking desktop must boot to a log in screen where it is possible to log in to a working desktop using a user account created during installation or a 'first boot' utility." 16:47:34 ack 16:47:35 ack 16:47:36 is that too long? 16:47:39 ack 16:47:39 ack 16:47:41 ack 16:47:56 #agreed 1004902 - AcceptedBlocker - While not a 100% clear-cut violation of the criteria for KDE, this was deemed close enough to a violation to take as a release blocking bug for F20 alpha: "A system installed with a release-blocking desktop must boot to a log in screen where it is possible to log in to a working desktop using a user account created during installation or a 'first boot' utility." 16:47:58 ack 16:48:13 While not a 100% clear-cut violation lol 16:48:28 Viking-Ice: glad to see you like my wording 16:48:38 ok, that's all of the proposed blockers on my list 16:48:48 ack 16:48:57 oh, too late :) 16:49:04 moving on to the single proposed FE 16:49:11 #topic (1005435) Trimslice requires kernel-3.11.0-300.fc20 for network in F20 Alpha 16:49:14 #link https://bugzilla.redhat.com/show_bug.cgi?id=1005435 16:49:16 #info Proposed Freeze Exceptions, kernel, ON_QA 16:49:20 +1 16:49:23 +1 16:49:25 +1 16:49:29 +1 16:49:31 +1 16:49:32 +1 16:50:25 +1 16:50:29 proposed #agreed 1005435 - AcceptedFreezeException - This fixes wired networking on the trimslice which is a supported platform for F20 alpha. A tested fix would be considered after freeze 16:50:30 ack 16:50:36 ack 16:50:37 ack 16:50:38 ack 16:50:40 ack 16:50:46 #agreed 1005435 - AcceptedFreezeException - This fixes wired networking on the trimslice which is a supported platform for F20 alpha. A tested fix would be considered after freeze 16:51:11 now for the not {post,modified,on_qa} accepted blockers 16:51:17 of which, I see 1 16:51:26 great :) 16:51:26 #topic (997690) SizeNotPositiveError: bytes= param must be >=0 16:51:26 #link https://bugzilla.redhat.com/show_bug.cgi?id=997690 16:51:26 #info Accepted Blocker, anaconda, ASSIGNED 16:51:36 hi 16:52:24 looks like we're waiting for info on this one 16:53:21 so let's move on 16:53:50 #info this was supposed to be fixed but new reports have surfaced in the last several days 16:54:00 is reclaim space still considered as autopart? 16:54:03 #info new tb is needed for this to be fixed again 16:54:09 jreznik: AFAIK, yes 16:54:25 yes 16:54:33 ok, anyone from here seeing the bug and can provide required info? 16:54:54 Amit is in australia, so I doubt he's online right now 16:54:59 (or try to reproduce it) 16:55:16 I've yet to see this one, myself 16:55:36 satellit_e: how often are you seeing 997690? 16:55:40 satellit_e: I see you was able to hit too 16:55:47 I did see it when testing earlier...but not lately 16:56:05 so it might be fixed 16:56:41 yeah, both reports are with TC4 16:57:21 TC5 DVD installs to VirtualBox now default install 16:57:27 either way, needs more data. if nobody else can hit this, we can look at reclosing 16:57:35 so satellit_e reports it's possibly fixed, so ask amit if he can retest 16:57:41 tflink: yep 16:58:05 anything else on this bug? 16:59:00 * tflink takes silence as "no" 16:59:21 are there any accepted blockers that people want to discuss? 17:00:15 * dan408 is looking 17:00:25 who's the one with that late review cropping up now 17:00:39 ? 17:00:45 refresh the list 17:01:04 it should block the tracker, not depends on 17:01:21 and https://bugzilla.redhat.com/show_bug.cgi?id=1005906 should be added there too 17:01:53 package review this late in the process wtf some needs a slap at the back of the head for this 17:02:33 Viking-Ice: elections were too late... 17:02:33 * dan408 doesnt have anything 17:02:50 blocked on legal, then issues with announcement... 17:03:08 knew that adding different wallpaper to block the release would be an issue 17:03:16 #topic Open Floor 17:03:50 Viking-Ice: there is an oddity with how their workflow is going but I'm not seeing the package review bugs on the list right now 17:04:19 Any other topic to bring up? 17:04:29 https://bugzilla.redhat.com/show_bug.cgi?id=1004421#c3 not a blocker 17:05:21 http://wiki.sugarlabs.org/go/Fedora_20#SoaS_f20-Alpha_TC5 17:05:40 tflink: there's no review bug, package is under development 17:06:29 satellit_e: do you know if anyone is working on it? 17:06:54 I have pm'd pbrobinson he did last fix to .ks 17:07:41 pbrobinson is on vacation 17:07:49 may be slow to respond 17:07:50 ok, so for now - we really need to finish wp packaging asap, kde theme added as blocker for the bug, I'll ask martin to do so for heisenbug-backgrounds too 17:08:00 so we will know where we are 17:08:17 yeah, it'd be good to get a RC today or tomorrow 17:09:31 if there's nothing else, I'm setting the fuse for (0,5] minutes 17:09:34 tflink: backgrounds should be later today, I'm mostly done with kde-theme review (waiting for real backgrounds package) and we need kde-settings... I'll ask rex... so maybe it would be doable today 17:10:03 we're also waiting on some anaconda stuff 17:10:27 the i386 fix, more data on the data reclaim bug 17:11:03 * satellit_e I cannot install to ext USB HD from latest anaconda in Desktop TC5 ? hangs on reclaim space 17:11:10 hmm 17:11:27 satellit_e: could you attach logs? 17:11:33 usb2 or usb3 ports 17:11:37 as requested by dlehman 17:11:38 are getting a TC6 soon or an RC1? 17:11:42 jw 17:11:46 dan408: see above 17:11:48 will try 17:12:02 ok 17:12:06 RC got it 17:12:19 just trying to keep up 17:12:20 dan408: it depends :) 17:12:28 depends on what? :P 17:12:40 sorry im so behind 17:12:50 dan408: on if all the blockers are addressed 17:12:56 k 17:12:59 if there are still unaddressed blockers, no RC 17:13:27 okay someone tell me do we need mote dvd testing or netinstall testing? 17:13:39 i have more blank dvds than cds 17:13:43 both? 17:13:55 and is custom partitioning fixed? 17:13:55 I don't see any optical media DVD tests yet, though 17:14:03 dan408: not that I'm aware of 17:14:13 still to big for DVD 17:14:17 yeah i just hate wasting them to see another spinup the next day 17:14:21 too big for dvd??? 17:14:23 ok guys, I have to go no, will be online later today to coordinate wp bug 17:14:30 cya jreznik 17:14:41 jreznik: cool, see you in a bit :) 17:14:55 * satellit_e afk.... 17:15:01 * tflink notes that the fuse is getting dangerously close to the end ... 17:15:16 * dan408 woulld like to propose that all images go here http://dl.fedoraproject.org/pub/alt/stage/20-Alpha-TC5/Images/x86_64/ 17:15:24 we have too many directories and it's annoying 17:15:40 dan408: you'd have to talk to RelEng about that 17:15:42 * dan408 waits to get voted down 17:15:56 okay so the DVD is oversized by 200MB 17:16:53 ill try and do me part with MATE today that might shave off 100MB if im lucky 17:17:07 thus only VirtualBox works last time I tested dd USB 8 GB it could find it /dev/sr0 only available 17:17:16 tflink: dont you think the DVD being oversized is a blocker? 17:17:20 it could not* 17:17:24 I mean it's no longer a blocker 17:17:27 err 17:17:30 no longer a dvd 17:17:47 it's a DL DVD now 17:17:48 heh 17:18:29 ok, I think the restof this can be moved to #fedora-qa 17:18:30 well i guess ill be testing a netinstall image i guess 17:18:41 thanks for coming, everyone! 17:18:47 * tflink will send out minutes shortly 17:18:50 #endmeeting