17:00:34 #startmeeting F20-blocker-review 17:00:34 Meeting started Wed Nov 20 17:00:34 2013 UTC. The chair is pschindl. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:34 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:00:38 #meetingname F20-blocker-review 17:00:38 The meeting name has been set to 'f20-blocker-review' 17:00:42 #topic Roll Call 17:00:45 bang! 17:00:48 * kparal is here 17:00:51 * roshi is here 17:01:00 * pwhalen is here 17:01:14 * Viking-Ice fetches coffee not for you guys for myself ;) 17:02:25 Hi guys. 17:02:38 Is anyone else coming? 17:02:44 * mkrizek lurks 17:03:21 Ok. Let's start. There is lot of bugs before us. 17:03:26 ahoyhoy 17:03:30 #topic Introduction 17:03:38 Why are we here? 17:03:44 #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 nice-to-have bugs. 17:03:47 #info We'll be following the process outlined at: 17:03:49 #link https://fedoraproject.org/wiki/QA:SOP_Blocker_Bug_Meeting 17:03:51 #info The bugs up for review today are available at: 17:03:53 #link http://qa.fedoraproject.org/blockerbugs/current 17:03:55 #info The criteria for release blocking bugs can be found at: 17:03:57 #link https://fedoraproject.org/wiki/Fedora_20_Alpha_Release_Criteria 17:03:59 #link https://fedoraproject.org/wiki/Fedora_20_Beta_Release_Criteria 17:04:01 #link https://fedoraproject.org/wiki/Fedora_20_Final_Release_Criteria 17:04:07 What is waiting for us? 17:04:12 #info 9 Proposed Blockers 17:04:15 #info 14 Accepted Blockers 17:04:17 #info 2 Proposed Freeze Exceptions 17:04:19 #info 4 Accepted Freeze Exceptions 17:04:42 Who is willing to do secretary work today? 17:04:48 * adamw will 17:05:03 adamw: Thank you very much :) 17:05:39 So. Lets take a look on proposed blockers 17:05:45 #topic (1029616) ValueError: ('invalid size specification', '0.00 MB') 17:05:47 #link https://bugzilla.redhat.com/show_bug.cgi?id=1029616 17:05:49 #info Proposed Blocker, anaconda, NEW 17:06:31 Does somebody knows which languages break the thing? 17:06:33 argh language bug still! 17:06:36 so i'm not totally sure of the conditions to trigger this yet, but it's a complete showstopper if you hit it 17:06:43 +1 blocker 17:06:44 pschindl: so far I think we know of French and Kazakh 17:06:51 but i'll have to go through the whole list and see which are broken 17:07:04 adamw: good luck 17:07:18 yeah, i love that testing 17:07:25 I've tried two languages and both worked fine 17:07:40 as for comment 24, Czech also uses comma for decimals and it seems to work fine 17:08:04 that's my observation too 17:08:27 so, it's not that, but something else... 17:08:28 votes on blocking 17:08:32 +1 blocker 17:08:34 so, this is related to the language chosen in anaconda welcome screen, or the language of previous installation? 17:08:34 but still. I'm +1 for blocker 17:08:48 my inclination is +1 just for hitting french, but i'm happy to try and provide more data if folks want it 17:08:54 kparal: language chosen in welcome screen 17:09:02 kparal: if you run the installer again and pick english it works fine 17:09:18 +1 blocker 17:10:33 the criterion doesn't seem to be included, no cookie for adamw 17:10:41 aw! 17:10:50 adamw: you're on a diet lately! 17:11:02 conditional violation of "The installer must run when launched normally from the release-blocking images." 17:11:12 well, i suppose it runs. 17:11:19 so, "When using the dedicated installer images, the installer must be able to complete an installation using the text, graphical and VNC installation interfaces. " 17:11:25 no. grr. damnit. 17:11:51 why does that only say dedicated installer images? what the hell kind of crack was that guy who wrote it on>? 17:12:03 the first is fine, it doesn't run exactly, crashes right after start 17:12:27 yeah, let's call it that. 17:12:39 adamw: the dedicated word is there probably because of vnc 17:12:40 though i really had 'it doesn't run at all' in mind when i was writing it, but eh. 17:12:44 kparal: yeah, indeed 17:12:48 i need to re-work that 17:13:07 proposed #agreed 1029616 - AcceptedBlocker - Anaconda crashes when some languages are choosen for installation (for example French). This violates the criterion: "The installer must run when launched normally from the release-blocking images." 17:13:10 ack 17:13:16 ack 17:13:17 ack 17:13:27 +1, ack 17:13:37 #agreed 1029616 - AcceptedBlocker - Anaconda crashes when some languages are choosen for installation (for example French). This violates the criterion: "The installer must run when launched normally from the release-blocking images." 17:13:56 #topic (1032620) Doesn't remember POP password 17:13:59 #link https://bugzilla.redhat.com/show_bug.cgi?id=1032620 17:14:01 #info Proposed Blocker, evolution, NEW 17:14:23 tflink: blockerbugs app down 17:14:39 eh? 17:14:41 works fine here 17:14:47 I see http 503 17:15:02 and it's back up! 17:15:06 your employeer wants you to focus your time on RHEL now 17:15:12 lol 17:15:16 * adamw runs screaming 17:15:18 :D 17:15:39 that proposed bug seems to be a criteria autoblocker 17:15:46 tflink: maybe check the logs, but it might have been a problem with my browser or something. still http 503 seems weird 17:15:52 +1 17:16:02 +1 17:16:06 no wait pop not working that's a feature 17:16:14 evolution is not on LiveCD, is it? 17:16:15 Does someone else met this? 17:16:30 yeah...i'd be +1 but seems odd no-one else has hit it 17:16:30 I use evolution and I haven't seen such a problem 17:16:40 does anyone still use pop these days ;) 17:16:43 pschindl: you use imap 17:16:55 he says it affects calendars as wel 17:17:10 I use evolution, but for imap - I can test with pop today 17:17:21 we should probably test it in a clean from-VM install 17:17:31 is evolution in the default install? 17:17:36 I wonder if it hits when you add an online account, like during initial setup 17:17:37 ah, ok, I thought that one of my accounts is pop, but no, both are imap. Sry :) 17:17:38 kparal: yeah, found it 17:17:43 kparal, yeah think so 17:17:51 how can I check quickly? in comps.xml? 17:17:52 I suspect it's an issue with load balancing and out change to openid 17:17:53 it's in the default 17:18:06 kparal: evo is on the live cd, yeah. 17:18:12 alright 17:18:15 punt for testing/confirmation ? 17:18:35 how about we vote that it will be a blocker if it can be reproduced? 17:18:38 and action someone to reproduce? 17:18:52 I've already voted just waiting on you guyes ;) 17:19:09 well the cited criterion is not exact. it might be a problem with evolution, not gnome-keyring 17:19:30 this covers it better: http://fedoraproject.org/wiki/Fedora_20_Final_Release_Criteria#Default_application_functionality 17:19:45 POP is still used a lot, I think 17:19:47 kparal: eh, i think the keyring thing is correct 17:20:00 that _is_ what evo uses to store passwords 17:20:01 so votes people 17:20:06 dissect later 17:20:16 provisional +1 if it can be reproduced by someone else from a fresh install 17:20:23 +1 blocker if reproducable 17:20:26 +1 17:20:27 in any other case, we'd have to re-discuss 17:20:38 +1 provisional 17:20:40 * kparal agrees 17:21:13 ok. So who will test it? 17:21:14 what is our approach with conditional blockers? do we accept them right now? 17:21:26 or after someone tests it? 17:21:27 i'd say something like: 17:21:29 That's my question too :) 17:21:41 #chair adamw kparal 17:21:41 Current chairs: adamw kparal pschindl 17:22:11 proposed #agreed #1032620 would be accepted as a blocker so long as it can be reproduced from a fresh install by someone else. roshi will attempt to reproduce and mark as accepted if so 17:22:14 #chair Viking-Ice 17:22:14 Current chairs: Viking-Ice adamw kparal pschindl 17:22:16 ack 17:22:16 * adamw throws roshi to the wolves 17:22:25 mmmm, wolf 17:22:27 poor wolves 17:22:29 hehe 17:22:41 ack 17:22:47 ack 17:22:48 good with some creole seasoning, a light ale and some sourdough bread 17:22:49 ack 17:22:50 and then roshi could try to reproduce, if he can, mark it as accepted, if not, just add a comment to that effect and it'll come up for discussion again next time 17:23:00 works for me 17:23:07 #agreed #1032620 would be accepted as a blocker so long as it can be reproduced from a fresh install by someone else. roshi will attempt to reproduce and mark as accepted if so 17:23:10 animal cruelty 17:23:34 #topic (1013280) Slideshow in Anaconda 20.21 is only in Enlish language while in other distros like Ubuntu the language of the slideshow in the installer is according to the selected language for installation 17:23:37 #link https://bugzilla.redhat.com/show_bug.cgi?id=1013280 17:23:39 #info Proposed Blocker, fedora-logos, MODIFIED 17:23:40 what, for having to eat me or being eaten? 17:24:20 so, this is not about release notes but marketing materials - banners 17:24:42 -1 17:24:45 well, i think it covers more than just what the topic says 17:24:45 is this something to take to the marketing team then? 17:24:51 Why that should be blocker? 17:24:55 adamw: I'm referring to your blocker proposal 17:25:14 other languages don't show in anaconda because there are no translations of the banners, or because of bug in anaconda? 17:25:29 i meant that i thought spot asked me to add the bug because it includes actual release notes 17:25:34 adamw, good luck for marketing translating it to Icelandic muhahah 17:25:42 but now i read the update notice again, 'rnotes' could be the name of the directory with the banner images in it 17:25:51 those fracking ad banner should be dropped ( unless we get paid for them, then we should keep them ) 17:26:05 * adamw checks 17:26:38 ah, yeah, i think it is talking about the ad banners 17:26:39 if there is a bug in anaconda _preventing_ localized banners to be shown, then we can vote about it. if there are simply no translations available, then -1 17:26:55 so i guess i withdraw blocker nomination and make it FE instead 17:27:21 -1 / +1 FE 17:27:21 the update is already pushed to testing, so it might make it before the freeze 17:27:23 kparal: it's neither of those: there are translated images available but we simply didn't put them in the package. the update that fixes this bug is a package that includes translated images. 17:27:23 +1 FE this doesn't seem to be blocker. 17:27:28 -1/+1 FE 17:27:44 yeah, it could make stable before freeze, but may as well vote it an FE now so we can pull it in if not. 17:27:46 -1/+1 17:27:48 -1/+1 17:27:50 I see. -1/+1 17:28:35 BTW, how big are those images ? 17:28:59 how big can you carry ;) 17:29:29 mkolman: 10MB package 17:29:31 proposed #agreed 1013280 - RejectedBlocker AcceptedFreezeException - Translation in ad banners doesn't break any criteria. But it's worth to have possibility of having translated banners. 17:29:33 mkolman: good question, but not too big. 17:29:37 could be an issue as the installation is basically running from a ramdisk 17:29:52 well what i'd be more worried about is making the live images oversize again, actually...better check that 17:29:52 well, both how many translations are that ? 17:29:55 ack 17:30:02 ack 17:30:03 ack (but let's check the size question too) 17:30:07 ack 17:30:47 if the banners are not svg, we probably need to go for svg in the long run 17:31:03 raster graphics take a lot of space in 40 languages 17:31:12 proposed #agreed 1013280 - RejectedBlocker AcceptedFreezeException - Translation in ad banners doesn't break any criteria. But it's worth to have possibility of having translated banners. The size of images have to be checked after this patch is pushed. 17:31:28 I added the last sentence. 17:31:30 ack 17:31:33 exactly, that's why I was asking 17:31:36 ack 17:31:37 ack 17:31:45 ack 17:31:51 #agreed 1013280 - RejectedBlocker AcceptedFreezeException - Translation in ad banners doesn't break any criteria. But it's worth to have possibility of having translated banners. The size of images have to be checked after this patch is pushed. 17:31:59 on the toher hand, svg should work just fine on the squashfs images 17:32:12 #topic (960381) gnome-shell eats a lot of mem 17:32:15 #link https://bugzilla.redhat.com/show_bug.cgi?id=960381 17:32:17 #info Proposed Blocker, gnome-shell, NEW 17:33:46 we need someone to reproduce this with a clean install. it can be caused by an extension, for example 17:34:37 I haven't seen gnome-shell consuming so much memory. 17:34:40 I don't see any problem myself 17:34:53 not seeing it here running but that might because that gnome-shell res mem in that report pails in comparision with thunderbird memory usage ( 1.4G ) 17:35:08 lol let's punt for more data 17:35:15 * adamw up 1 day 18 hours, shell using 505MB 17:35:31 yeah, i'd say punt 17:35:41 shell for me maxes out around 600MB 17:35:45 +1 punt 17:35:47 I'm up 3 days and I have 500MB usage for gnome-shell 17:36:01 I'm -1 unless easily reproduced on a clean install 17:36:25 I'm +1 punt. 17:36:36 this can be a witch hunt, can be caused by so many factors 17:36:53 yeah like Tim hacking cups ;) 17:36:54 yeah, 'memory leaks' can get that way 17:37:09 but i'd expect that twaugh should be capable of nailing it down if he sits down and tries 17:37:38 unless it's really easy to reproduce or debug, we might not fix this in time anyway 17:38:04 I'm down $100 if Tim prints out the debug output and smack the relevant maintainer in the head with 17:38:10 it 17:38:13 lol 17:38:31 * adamw asks in #fedora-desktop 17:39:28 nothin' yet, let's just punt it 17:39:57 ask Tim to add some debug info 17:40:04 yeah 17:41:48 #info This bug needs more info to have good decision. We will discuss it next time 17:42:12 is it enough for punt? 17:42:14 need to go out smoke -1 on " Switching between messages in message-tray doesn't work" 17:42:34 #topic (1031530) Switching between messages in message-tray doesn't work 17:42:37 #link https://bugzilla.redhat.com/show_bug.cgi?id=1031530 17:42:39 #info Proposed Blocker, gnome-shell, NEW 17:43:36 this might be annoying for empathy users 17:43:46 lots of notifications in the tray 17:44:08 -1 blocker 17:44:11 still I don't think it's a blocker 17:44:14 yeah 17:44:18 i can see where it hits the criteria, but 'common sense' seems to be -1 17:44:19 FE maybe though 17:45:21 this is one of the old 'strict polish' desktop criteria...not sure i'd want to lose it entirely, but...ehhh. i hate criteria logic. 17:46:43 anyhow, i'm -1 17:46:51 Ok. I agree that it's not so bad to block release. But it makes me crazy all the time I use it. And I use it quite a lot. Especially for chatting. 17:47:01 So I'm at least +1 for FE 17:47:34 * kparal does not like to waste time voting on FE when the developers haven't shown intention to actually fix it soon 17:47:48 let's vote on FE once there is a patch ready 17:48:07 makes sense kparal 17:48:30 the problem with that is the timing - if the bug is acceptedFE we can push it into an RC, get it tested, and push it stable fast 17:48:53 if we wait for the fix to be written before voting we introduce up to a possible 4 day delay in that loop (wednesday -> monday) 17:49:00 in that case we need developer->QA communication which would tell us which bugs to vote on in advance 17:49:10 say an update that fixes it arrives two minutes after this meeting finishes 17:49:20 we would not be able to pull it into an RC or push it stable until next monday at the earliest 17:49:25 (if we were frozen now, obviously) 17:49:40 we can't vote on every bug in bugzilla 17:49:56 this is just a random bug 17:50:27 we need to know dev's priorities so that we don't waste time voting on bugs that are not going to be fixed 17:50:27 proposed #agreed 1031530 - RejectedBlocker - Even though the notification area doesn't work properly basic funtionality is working well. There is no need to block release. 17:50:35 ack 17:50:36 ack 17:50:44 ack 17:50:48 ack 17:50:57 #agreed 1031530 - RejectedBlocker - Even though the notification area doesn't work properly basic funtionality is working well. There is no need to block release. 17:51:07 #topic (1031848) xkb-converted layouts do not include multiple mappings where they are commonly expected 17:51:10 #link https://bugzilla.redhat.com/show_bug.cgi?id=1031848 17:51:12 #info Proposed Blocker, kbd, NEW 17:51:29 come join me in my rabbit hole, folks 17:51:36 this one is so fun. many kudos to adamw for describing the situation clearly 17:51:51 might be a bit long to read, though 17:52:10 adamw: maybe tl;dr version for the rest? 17:52:15 adamw, did you fix this ( assuming the systemd bug that this got discussion on is related ) 17:52:41 Viking-Ice: i have an approach which seems to fix it in my testing, but i'd need vitezslav to sign off on it 17:53:07 so, tl;dr...er. 17:53:26 tl;dr it is a blocker and don't care :) ? 17:53:32 so what's the risk here ( thinking encrypted passwords ) 17:53:35 pretend you're Russian, okay? the way you expect your keyboard to work is that you can switch the layout 17:54:00 if you press 'a', you will get either an 'a' or a 'ะค' depending on which mode your keyboard is switched into 17:54:25 the way we have things set up in f20 at the moment, that doesn't work at the console 17:54:38 but used to work in F19, IIUIC 17:54:39 if you install in russian, you can only input cyrillic characters at the console. not latin. 17:54:55 adamw: which means you can't even log in? 17:55:01 yes, this is due to the change to xkb-converted layouts for kbd in f20, which i'm trying to not explain because it would make things longer :) 17:55:11 kparal: yup, basically, the console is utterly unusable 17:55:15 +1 blocker all things for mother Russia 'Zazdarovje' 17:55:38 it doesn't just affect russian - it affects any case where you pick a layout which requires this kind of switching to work 17:56:01 +1 blocker 17:56:06 like cz-us-qwertz. we know it 17:56:07 +1 blocker 17:56:09 that means any kind of cyrillic layout and a lot of indian/pakistani ones and arabic 17:56:41 kparal: czech is actually rather...odd in the mappings, i haven't tested how it actually winds up behaving yet. i should probably check 17:56:55 adamw: the fix is that you delete those systemd-generated layouts which overlap the previous console layouts and cause problems? 17:57:08 kparal: it's not just ones that overlap 17:57:21 adamw: it's very odd indeed. that's why no sane person uses it. certainly not in console :) 17:57:37 * kparal might be wrong, there are lots of lunatics out there 17:57:41 the intent in f20 was that everyone should use xkb-converted mappings, the legacy ones were only included for safety 17:57:53 but when we came up with that plan we didn't entirely think about this wrinkle 17:58:13 so, anyway, my approach to fix it is simply to delete all the xkb-converted layouts which aren't capable of inputting ASCII, since there's really no reason anyone would ever want to use such a thing 17:58:15 what was the code named of the plan 17:58:26 and the safest way to make sure they most definitely never wind up getting used is simply to get rid of 'em :) 17:58:49 an associated change to kbd makes the legacy layouts available again (but after the xkb-converted layouts in the search path) 17:58:57 in anycase votes I think that essay in the report was enough to xplain the situation 17:59:00 adamw: I'm just a bit nervous about grepping the layouts and finding "A" --> can input ascii. was that improved? 17:59:19 kparal: having spent all day on this yesterday i actually think that's the best approach 17:59:35 proposed #agreed 1031848 - AcceptedBlocker - Impossibility of changing layout in console makes console unusable for some layouts (for example russion). 17:59:39 kparal: the alternative is to use langtable, which is just a manually populated list so it's not necessarily better, and it's certainly badly incomplete. and it is more complex to implement 17:59:47 kparal: but if you have a better idea...i'm all ears 18:00:00 pschindl: Russian 18:00:01 kparal: fwiw i tested with various latin characters and you get the same result whichever you use (162 layouts match) 18:00:06 * handsome_pirate shows up late 18:00:20 and I built a live image with the patched systemd and kbd and actually tested UK english, Russian and Arabic installs, and they all give a sane result 18:00:26 ack 18:00:27 i'll be doing more testing, though. including czech. ;P 18:00:37 ack with s/russion/russian/ 18:00:55 adamw: I could help with that, I suppose. into the rabbit hole. or... instruct interns! 18:01:08 any other ack/nack/patch? 18:01:12 kparal: i'd have to stick the live image up somewhere...or you could build your own 18:01:16 kparal: Sounds like a good job for interns 18:01:46 ack 18:01:58 kparal: all you need is https://admin.fedoraproject.org/updates/systemd-208-6.fc20 and the scratch kbd I posted in https://bugzilla.redhat.com/show_bug.cgi?id=1031848#c9 18:02:01 #agreed 1031848 - AcceptedBlocker - Impossibility of changing layout in console makes console unusable for some layouts (for example russian). 18:02:21 #topic (1026860) Instantiated service is not run, it stays in inactive state (and systemd debug log does not state why) 18:02:23 #link https://bugzilla.redhat.com/show_bug.cgi?id=1026860 18:02:25 #info Proposed Blocker, lvm2, NEW 18:04:04 +1 18:05:03 from what I see, it seems to be a race condition for LVM on RAID 18:05:21 * handsome_pirate is +1 blocker 18:05:34 +1 18:06:07 * adamw reading 18:06:14 omg 18:06:51 +1 i guess 18:06:55 Viking-Ice: ? 18:07:08 * handsome_pirate is slipping 18:07:15 Only nine terminal windows open 18:07:18 I think I've done some raid tests (with lvm as a default partitioning scheme) and haven't seen this 18:07:38 it's a question how frequent this is 18:07:50 kparal, seems to be a race scenario 18:08:07 yeah. still it might not affect everybody 18:08:13 ? 18:08:23 so I guess +1 if it is generally affecting every LVM on RAID 18:08:42 if it affects only certain configurations, then it might not be a blocker 18:09:03 but it might be hard to find that out 18:09:14 it doesn't hit every lvm-on-raid config, no. i tested that for beta and it was fine. this bug actually developed out of the one i hit with earlier beta builds. 18:09:22 but at least two people have hit it. 18:09:33 could be pretty hard to guess how widespread it'd be in The Real World. 18:09:56 even harder since it's an race condition so some people might hit this while others do not 18:10:15 that's what I mean 18:10:20 it's not just 'everyone who does lvm on raid' 18:10:46 hence let's not risk it throw it on the list and here from Peter next meeting 18:10:52 mean hear 18:11:33 i'd be okay to punt, i guess 18:12:10 the problem is still being debugged, I'm OK with punt as well 18:13:24 #info There are still some information missing. We will try to get more information and return to this bug next meeting 18:13:27 ack 18:13:37 dont ask me why I acked 18:13:41 ;) 18:14:12 That's good I'm glad that someone is acking what I write :) 18:14:14 ack 18:14:17 =) 18:14:22 #topic (1026283) Nautilus eating 100% cpu after kinit 18:14:25 #link https://bugzilla.redhat.com/show_bug.cgi?id=1026283 18:14:27 #info Proposed Blocker, nautilus, NEW 18:15:03 spoiler alert: pschindl have just reproduced this 18:15:05 I've met this one three times during last week. 18:15:57 pschindl: have you run kinit always before seeing this? 18:16:18 which actions run kinit? 18:16:42 pschindl: it's a command you run in bash. or maybe you have kerberos configured in online accounts 18:16:45 if you have remote auth set up then it gets done when you log in... 18:16:53 right, or in GOA 18:17:36 I haven't done anything like that. 18:17:52 I don't have set kerberos account. 18:18:02 might be a different bug then 18:18:33 maybe. But it causes same symptoms 18:18:55 oh hey, i see some discussion in #fedora-desktop 18:19:04 * kparal peers 18:19:08 rishi: hey, I see you found https://bugzilla.redhat.com/show_bug.cgi?id=1026283 ? 18:19:08 Bug 1026283: high, unspecified, ---, mclasen, NEW , Nautilus eating 100% cpu after kinit 18:19:14 mclasen: Yes, I saw it in the backlog. 18:19:20 mclasen: That kerberos code is turning out to be a never ending saga. 18:19:20 I wonder if it is related to the keyring cache in F20. 18:19:28 #link https://bugzilla.gnome.org/show_bug.cgi?id=710413 18:19:30 could be the same as bug 710413 18:19:30 Bug http://bugzilla.gnome.org/show_bug.cgi?id=710413 major, Normal, ---, nautilus-maint, UNCONFIRMED, nautilus process sometimes uses 100% cpu 18:19:40 to be honest, i don't have any reason to believe the gnome-online-accounts kerberos code is related at all 18:19:40 it seems pretty far fetched 18:19:40 though it's hard to say for sure without stacktraces 18:19:59 Viking-Ice: fwiw you don't have to bother with '#link' - meetbot picks up URLs automatically for the summary 18:20:10 dam his smartness ;) 18:21:14 do you have icons-on-the-desktop enabled? 18:21:23 pschindl: ^^ 18:21:25 * adamw trying to figure out why this nautilus process exists at all, he doesn't have one 18:21:28 adamw: no 18:21:42 pschindl: but you have a nautilus window opened? 18:21:46 no :) 18:21:53 i'd say View With Concern but not +1 till we have more details on what's going on 18:22:21 so punt 18:22:28 yep punt 18:22:31 yes, I'm not +1 right now. I think it would be better to debug it firstly 18:23:02 But I think it could be blocker then. :) 18:23:20 +1 punt 18:24:06 yeah, puint 18:24:32 we might want to head over to meeting for a second ( lvm-thinp/btrfs discussion ) 18:25:07 Viking-Ice: #fedora-meeting? 18:25:19 yep 18:25:30 #info It is not clear what causes the problem and how big the problem is. We will discuss it next meeting after (hopefully) getting more information 18:25:40 can we do that in parallel? 18:26:01 probably best pause this meeting for a bit 18:26:54 we can throw in the punt and ack on it I suppose 18:27:14 ok kids you have ten minutes. But then we will continue in our class :) 18:28:12 Viking-Ice: info just made the punt thing, so we can now focus on #fedora-meeting 18:42:02 wow, those fesco meetings are messy 18:51:48 yup. 18:51:51 looks like we can move on now 18:52:01 #topic (1006386) Boot takes 27 seconds longer with /var/log/journal than without 18:52:03 #link https://bugzilla.redhat.com/show_bug.cgi?id=1006386 18:52:05 #info Proposed Blocker, systemd, NEW 18:54:06 any votes? Notes? Something? 18:54:15 Is here anybody? :) 18:54:31 sorry, was catching up with the last one 18:55:06 still reading 18:55:38 looks like we need to be reading https://bugzilla.redhat.com/show_bug.cgi?id=1006386#c39 18:56:17 the problem is that the original reporter says nothing of broken gdm 18:56:39 robatino seem no gdm in 50% of boots 18:56:47 might be a completely different issue 18:56:59 the OP just experienced 30sec delayed boot 18:57:00 someone mentions 'at least three different issues' 18:57:05 yeah 18:57:16 "yeah, it got pretty messed, basically at least three issues that should have been tracked as three separate bugs ..." 18:57:17 so I'm not really sure what we vote on 18:57:22 "1) slow journal blocking other things - let this one be primary bug for this" 18:58:07 ah, "It is NOT "just making startup slow". Services often fail to start completely or hang in non-working state, so that kdm/gdm never comes up" 18:58:21 the title makes it sound like a non-blocker 18:58:34 * roshi hasn't seen this 18:58:36 see my previous comment 18:58:43 I've not seen this either 18:58:51 yeah - it's just an unfortunate title 18:59:03 so karel's contention appears to be that the bug really happens to everyone, but if your hardware is sufficiently fast, you don't hit the bad consequences 18:59:10 what kinda hardware are you guys on? mine is reasonably fast, indeed 18:59:24 mine is 18:59:31 but I don't get this on a netbook 18:59:34 I have SSD. but I don't experience it with F19 and HDD at home 18:59:35 t420 + ssd 18:59:44 either 18:59:50 1gb ram 1 ghz atom and a 54 spinny disk 18:59:56 do people still use rotating media 19:00:00 ? 19:00:03 * Viking-Ice hides 19:00:05 my netbook runs f20 fine 19:00:22 you do when it's $150 and is easily transportable :) 19:00:53 I say punt we ( by we I mean brno team ) can ping Michal 19:00:55 but, it is i686 not x86_64 19:00:59 Viking-Ice: my salary doesn't run to 10TB of SSDs, so my NAS still spins :) 19:01:24 michal seems to implicitly acknowledge the bug in c#51, but doesn't provide any of the 'how bad is it' evaluation we wnated 19:01:32 adamw, I actually just stream things these day's no storage 19:01:44 Viking-Ice: i don't really trust other places to store everything for me :P 19:01:57 if we were to just trust karel's description i'd be +1 19:02:07 but i'd still kinda like a systemd dev to confirm it 19:02:15 * kparal agrees 19:02:17 I concur ^^ 19:03:12 so punt? 19:03:19 +1 punt 19:03:29 +1 19:04:21 #info We need more information. Decision is left for the next meeting. We will gather more information. 19:04:34 yeah, though we need to start ragging the people we're punting to to make sure they get back to us 19:05:03 please ask Michal to describe the severity of this 19:05:04 needinfo request on michal ;) 19:05:55 yeah, done 19:06:28 ok, that was the last proposed blocker. So now we will move to proposed freeze exceptions 19:06:36 #topic (1031252) [abrt] empathy-3.10.2-1.fc20: _g_log_abort: Process /usr/libexec/empathy-chat was killed by signal 6 (SIGABRT) 19:06:39 #link https://bugzilla.redhat.com/show_bug.cgi?id=1031252 19:06:41 #info Proposed Freeze Exceptions, folks, ON_QA 19:07:00 this is already fixed and stable 19:07:17 +1 FE 19:07:27 already stable? if it's already pushed, no need to vote 19:07:39 aww I like to vote 19:07:45 yeah, it's going to be pushed in the next stable push, so let's move on 19:07:49 sorry viking :) 19:07:50 it just requested stable 19:07:59 ok, move on 19:08:04 * Viking-Ice re-tracks the vote 19:08:07 ah, sry. It's not stable yet 19:08:42 yeah, but it'll go in the next push. 19:08:51 adamw: yes 19:09:02 pschindl: let's move on 19:09:14 #info this bug is already fixed. Build waits for push to the stable 19:09:23 #topic (1031836) Excluding ARM from the arches broke compose of many ARM images for Fedora 20 Final TC1 19:09:25 #link https://bugzilla.redhat.com/show_bug.cgi?id=1031836 19:09:25 why are they shipping yumex 19:09:27 #info Proposed Freeze Exceptions, python-pexpect, MODIFIED 19:09:45 +1 FE 19:09:52 +1 19:10:47 +1 19:12:28 +1 (and i pulled the fix for this into TC2 since it's a no-brainer) 19:12:37 and we're not frozen yet , so i'm allowed to do that :P 19:13:30 proposed #agreed 1031836 - AcceptedFreezeException - Unexistence of python-pexpect-2.5.1-11.fc20 for ARM caused missing ARM builds. 19:13:41 ack/nack/PATCH? 19:14:17 ack 19:14:35 ack 19:14:42 ack 19:14:54 #agreed 1031836 - AcceptedFreezeException - Unexistence of python-pexpect-2.5.1-11.fc20 for ARM caused missing ARM builds. 19:15:35 ok. That's all from proposed bugs. And now for something completely different 19:15:53 Our favorit part: accepted blockers 19:16:10 #topic (1008732) LUKSError: luks device not configured 19:16:13 #link https://bugzilla.redhat.com/show_bug.cgi?id=1008732 19:16:15 #info Accepted Blocker, anaconda, NEW 19:17:39 the problem is that nobody is able to reproduce it any more 19:17:43 pschindl: can you? 19:18:03 I haven't seen it since that one try. 19:18:37 I will try to reproduce it with TC2. 19:18:59 let's leave this one as it is, and we can waive the blocker if nobody reproduces it in the remaining weeks 19:19:30 yeah, this one is icky 19:19:39 #action pschindl to try to reproduce this bug with TC2 19:19:45 i can go with the approach kparal suggested, but we probably shouldn't have accepted it in the first place... 19:20:33 #info nobody seems to be able to reproduce this bug. If no one reproduce it we will consider waiving the blocker status 19:20:51 #topic (1013586) SizeNotPositiveError: spec= param must be >=0 19:20:53 #link https://bugzilla.redhat.com/show_bug.cgi?id=1013586 19:20:55 #info Accepted Blocker, anaconda, NEW 19:23:02 needs dev attention 19:23:19 yeah 19:23:25 this one seems pretty clearly defined, just waiting for devs 19:23:47 So who is willing to ask him? 19:24:38 i don't think it's necessary at this point 19:24:46 they know the accepted blocker list is the todo list 19:24:47 ok. 19:24:55 not really much sense in wasting time saying 'hey, did you fix THIS one yet?' 19:24:58 That's true 19:25:58 #info This bug is waiting for dev attention. 19:26:11 #topic (1020974) incorrectly treats a disk with partially corrupt GPT as having no partition at all 19:26:14 #link https://bugzilla.redhat.com/show_bug.cgi?id=1020974 19:26:16 #info Accepted Blocker, anaconda, ASSIGNED 19:27:54 The same problem here. 19:29:26 #info Anaconda team has lot of work, this one is still on their list waiting for action. 19:29:44 yeah, we can blow through ones that are clear-cut quite quickly 19:29:52 #topic (1027947) ValueError: new size same as old size 19:29:55 #link https://bugzilla.redhat.com/show_bug.cgi?id=1027947 19:29:57 #info Accepted Blocker, anaconda, ASSIGNED 19:30:40 adamw: anaconda-20.25.8-1.fc20 is available in bodhi 19:31:00 (reply to comment 21) 19:32:30 #info There is a build which should solve the problem, but it looks like the problem is still present. 19:32:32 kparal: it'll be in tc2 19:32:38 oh, i see 19:32:50 i know it's available in bodhi, but i wanted to clarify the reporter didn't assume it was in tc1 19:33:10 I think he just updated anaconda on Live. but sure, let's ask for confirmation 19:34:04 #action pschindl or kparal to ask karal (also known as jsedlak) how he tested (which anaconda was used) 19:34:15 s/karal/karel :) 19:34:55 #topic (1027965) CreateException: Can't have a partition outside the disk! 19:34:58 #link https://bugzilla.redhat.com/show_bug.cgi?id=1027965 19:35:00 #info Accepted Blocker, anaconda, NEW 19:35:27 pschindl: don't you dare confusing others 19:35:45 pschindl enjoys giving our interns different names that nobody remembers 19:35:52 haha 19:36:29 But they listen to them. Isn't it great we have to times more interns for the same price :) 19:36:34 need dev attention 19:37:03 #info This bug is also waiting for devs attention. 19:37:14 #topic (1028110) LVMError: lvresize failed for root: running lvm lvresize --force -L 8712m fedora/root failed 19:37:17 #link https://bugzilla.redhat.com/show_bug.cgi?id=1028110 19:37:19 #info Accepted Blocker, anaconda, NEW 19:38:09 #info Another anaconda bug which is on very long todo list and waits for attention. 19:38:24 #topic (1028367) Invalid resize operation crashes 19:38:27 #link https://bugzilla.redhat.com/show_bug.cgi?id=1028367 19:38:29 #info Accepted Blocker, anaconda, NEW 19:39:05 I think they have to hate us :) 19:39:44 yup :) 19:39:46 #info Bug waits for dev attention. 19:39:58 #topic (864198) grubby fatal error updating grub.cfg when /boot is btrfs 19:40:01 #link https://bugzilla.redhat.com/show_bug.cgi?id=864198 19:40:03 #info Accepted Blocker, grubby, ASSIGNED 19:41:08 yup, just waiting on a fix, i think 19:41:33 Is Peter working on it? 19:41:58 So no need for reminder? 19:42:39 he ought to be :P i'm posting an ote on each bug to serve as a reminder 19:42:43 and i'll do a blocker status email this week 19:42:52 ok. 19:43:17 #info Work on fix in progress. 19:43:30 #topic (1028207) non US keyboard layouts not working at console 19:43:33 #link https://bugzilla.redhat.com/show_bug.cgi?id=1028207 19:43:35 #info Accepted Blocker, kbd, NEW 19:44:44 really what i'd like here is to get a new kbd and systemd intended to address this, 1031848 and https://bugzilla.redhat.com/show_bug.cgi?id=981805 all into tc3 19:44:46 So this seems to be fixed. It just needs more testing. Am I right? 19:44:48 and then re-evaluate 19:45:06 ok 19:45:08 well...the live images i've been building to test the changes seem to not hit this bug 19:45:20 but we never entirely explained what's going on, so it's hard to say for sure if the 'bug is fixed' 19:45:23 * kparal will ping vcrhonek tomorrow 19:45:33 probably best get the changed kbd and systemd out into tc3 and ask as many people as possible to try it 19:47:42 #info It's not clear what causes the bug, but it seems to be fixed (it doesn't occur on custom build). Will need further testing, especially when new kdb and systemd will be in game 19:47:49 game? 19:48:00 *kbd 19:48:23 what is the command for undo? 19:48:29 #undo 19:49:14 game - because nothing better came to my mind. I wanted to say that we will have to test it with new kbd and systemd 19:49:17 #undo 19:49:17 Removing item from minutes: 19:49:24 roshi: thanks 19:49:56 np - you'd do the same for me :) 19:50:05 #info It's not clear what causes the bug, but it seems to be fixed (it doesn't occur on custom build). Will need further testing, especially with new kdb and systemd 19:50:31 #topic (1009828) UEFI boot menu doesn't contain safe graphics mode 19:50:34 #link https://bugzilla.redhat.com/show_bug.cgi?id=1009828 19:50:36 #info Accepted Blocker, LiveCD, ON_QA 19:51:22 should be fixed in tc2, needs confirmation 19:51:29 especially since i wrote the fix, so it'll probably eat your babies instead 19:51:41 we will test it 19:52:02 but in tc2, every image whether booted bios or uefi should have a 'basic graphics' boot entry, and it should pass 'nomodeset' but not 'xdriver=vesa' 19:52:07 that's the intended state, as recommended by ajax 19:52:34 #info Bug should be fixed in TC2. Fix need confirmation (keep your babies in safety) 19:52:49 =) 19:53:08 #topic (1030719) tmux segfaults on start in 32-bit Fedora 20 TC1 install images due to libevent-2.0.so.5 being corrupt, causing anaconda to fail to start 19:53:11 #link https://bugzilla.redhat.com/show_bug.cgi?id=1030719 19:53:13 #info Accepted Blocker, pungi, NEW 19:55:32 so we decided to just fire the tc2 compose and see what happens 19:55:41 dgilmore's had connectivity issues lately so hasn't been able to try and figure out what happened to tc1 19:55:53 it may be a magic pixie bug, in which case it'll go away in tc2 and we can stop caring 19:57:14 #info We are waiting for TC2. We will see if the problem still persists 19:57:34 #topic (1027682) DeviceError: ('Cannot destroy non-leaf device', 'fedora_dhcp-29-193-root') 19:57:37 #link https://bugzilla.redhat.com/show_bug.cgi?id=1027682 19:57:39 #info Accepted Blocker, python-blivet, ON_QA 19:58:22 waiting for tc2 19:58:28 yup 19:58:54 #info This bug should be fixed in TC2 and waits for testing. 19:59:11 #topic (790339) [abrt] system-config-services-0.101.7-2.fc17: connection.py:630:call_blocking:DBusException: org.freedesktop.DBus.Error.UnknownMethod: Method "list_services" with signature "" on interface "org.fedoraproject.Config.Services.ServiceHerder" doesn't exist 19:59:14 #link https://bugzilla.redhat.com/show_bug.cgi?id=790339 19:59:16 #info Accepted Blocker, system-config-services, NEW 19:59:45 hum, thought i saw there was a build to fix this? 19:59:57 ah, yes. 19:59:59 There is a build which fixes it 20:00:28 changing status to VERIFIED 20:00:31 just needs one more karma 20:00:37 (and should probably check that bug lnie filed) 20:00:38 yes. 20:02:00 I think that Lili tried one service which normally do nothing so she thought that it's not working. But we should ask her. 20:02:44 OK 20:03:06 #info There is a fix. Update needs more karma to get to the stable 20:03:29 And the last one 20:03:32 #topic (1008965) mouse cursor sometimes disappears on login 20:03:34 #link https://bugzilla.redhat.com/show_bug.cgi?id=1008965 20:03:36 #info Accepted Blocker, xorg-x11-server, NEW 20:03:53 I have to say that I haven't seen this bug for very long time 20:04:16 i'm still seeing it regularly in tc1 testing 20:04:16 I haven't either 20:04:19 in a VM, though 20:06:06 So it still exists. Ok. And there is no fix. 20:07:44 What can we do now for moving on here? 20:08:30 seems like the devs are playing ping-pong 20:08:52 for now we can ask whot and hadess to get together and decide what they're doing a bout it 20:13:54 #info Bug still persists. Devs are 'playing ping-pong' and nothing seems to progress. We should try to persuade them to cooperate and come up with solution 20:15:34 ok. I think we can skip proposed FE. It's too late. Or does someone want to go through them? 20:16:02 Does anybody survived to this point? 20:16:25 * adamw still alive 20:16:30 but we're over time, should've stopped 15 mins ago 20:16:49 we already did proposed FE, though, and we don't usually 'do' accepted FE 20:16:52 There was pause for Fesco meeting :) 20:17:05 ok. That's great :) 20:17:06 * kparal votes for #endmeeting 20:17:26 yes! 20:17:39 #info Next meeting time - 17:00 UTC on 2013-11-27 20:17:47 aaaaand 20:17:52 #endmeeting