16:00:25 #startmeeting F22-blocker-review 16:00:25 Meeting started Mon May 4 16:00:25 2015 UTC. The chair is adamw. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:25 Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:00:26 #meetingname F22-blocker-review 16:00:26 The meeting name has been set to 'f22-blocker-review' 16:00:26 #topic Roll Call 16:00:32 ahoyhoy folks 16:00:38 * pschindl is here 16:01:22 who else is around to review some blockers? 16:02:03 * satellit listening but have to leave early 16:02:30 ahi 16:02:39 aqui, whatever 16:02:42 kparal: tflink: pwhalen: dgilmore: poke 16:02:57 * pwhalen is here 16:02:58 ow 16:03:31 finally! the Remote Pokinator Mk. 2 is successful! now the WORLD is at my mercy... 16:03:43 #chair pwhalen tflink 16:03:43 Current chairs: adamw pwhalen tflink 16:03:57 #topic Introduction 16:03:58 Why are we here? 16:03:58 #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. 16:03:58 #info We'll be following the process outlined at: 16:03:58 #link https://fedoraproject.org/wiki/QA:SOP_Blocker_Bug_Meeting 16:03:59 #info The bugs up for review today are available at: 16:04:01 #link http://qa.fedoraproject.org/blockerbugs/current 16:04:03 #info The criteria for release blocking bugs can be found at: 16:04:05 #link https://fedoraproject.org/wiki/Fedora_22_Alpha_Release_Criteria 16:04:07 #link https://fedoraproject.org/wiki/Fedora_22_Beta_Release_Criteria 16:04:09 #link https://fedoraproject.org/wiki/Fedora_22_Final_Release_Criteria 16:04:12 adamw: will pay attention as I can 16:04:18 thanks dgilmore 16:04:18 in releng meeting 16:04:29 #info 8 Proposed Blockers 16:04:29 #info 8 Accepted Blockers 16:04:29 #info 1 Proposed Freeze Exceptions 16:04:30 #info 2 Accepted Freeze Exceptions 16:04:56 who's up for secretary duty today? 16:05:05 * jreznik is here 16:06:14 I'll take it, adamw - already set up here. 16:06:33 thanks dan 16:06:34 hi jreznik 16:06:39 #info danofsatx will secretarialize today 16:06:57 onto the proposed blockers! 16:06:58 * kparal lurks 16:06:58 #topic (1193389) Support for chrony pool directive 16:06:58 #link https://bugzilla.redhat.com/show_bug.cgi?id=1193389 16:06:58 #info Proposed Blocker, anaconda, POST 16:08:18 yeah, I'm not sure this is blocker material 16:08:23 +1 FE, though 16:08:36 yeah, unless i missed something i don't see how this blocks. 16:08:37 +1 FE 16:09:10 +1 FE. 16:09:48 +1 FE, it might land before freeze, so academic discussion - but earlier, better 16:11:08 proposed #agreed 1193389 - RejectedBlocker AcceptedFreezeException - this does not violate any criterion so far as we're aware, but we would be willing to accept a tested fix during the freeze 16:11:15 ack 16:11:16 ack 16:11:18 ack 16:11:29 ack 16:11:39 #agreed 1193389 - RejectedBlocker AcceptedFreezeException - this does not violate any criterion so far as we're aware, but we would be willing to accept a tested fix during the freeze 16:11:47 #topic (1217411) F22 Desktop Final TC1 x86_64 German Version for installation no delete of previous SW possible 16:11:47 #link https://bugzilla.redhat.com/show_bug.cgi?id=1217411 16:11:48 #info Proposed Blocker, anaconda, POST 16:12:10 the naming on this bug is slightly confusing, what it means is that the Reclaim Space page doesn't work 16:13:05 it will be broken in any live non-English language with a translated size unit, at least i think that's the trigger. it's another bug related to the unicode stuff. 16:13:36 +1 16:13:47 +1 16:14:52 +1 16:14:58 +1 16:15:02 +1 16:15:45 proposed #agreed 1217411 - AcceptedBlocker - violates "When using the guided partitioning flow, the installer must be able to: ... Remove existing storage volumes to free up space, at the user's direction", in the case of an affected language 16:16:07 ack 16:16:19 ack 16:16:20 ack 16:17:15 #agreed 1217411 - AcceptedBlocker - violates "When using the guided partitioning flow, the installer must be able to: ... Remove existing storage volumes to free up space, at the user's direction", in the case of an affected language 16:17:23 #topic (1217438) CryptoError: Failed to activate device: Operation not permitted 16:17:23 #link https://bugzilla.redhat.com/show_bug.cgi?id=1217438 16:17:23 #info Proposed Blocker, anaconda, POST 16:18:02 per #c16 i'm a cynical -1/+1, but i wouldn't fight +1. 16:18:08 agreed on the "last blocker" test: -1/+1 16:20:22 hmm....tricky. But I'd go with the above....it really should give you another chance to unlock it instead of just crashing. 16:20:49 -1/+1 too. 16:21:40 -1/+1 but again, it would be nice to get it fixed 16:21:56 well, the patch is posted, so it probably will be 16:22:19 yeo 16:22:21 yep 16:22:37 proposed #agreed 1217411 - RejectedBlocker AcceptedFreezeException - we suspect this would get hand-waved if it was the last blocker on go/no-go day on the basis that you should just get it right, but we definitely would like to see it fixed 16:22:49 ack 16:23:04 ack 16:23:11 ack 16:23:45 #agreed 1217411 - RejectedBlocker AcceptedFreezeException - we suspect this would get hand-waved if it was the last blocker on go/no-go day on the basis that you should just get it right, but we definitely would like to see it fixed 16:23:53 #topic (1217504) UnicodeEncodeError: 'ascii' codec can't encode character u'\xfc' in position 9: ordinal not in range(128) 16:23:54 #link https://bugzilla.redhat.com/show_bug.cgi?id=1217504 16:23:54 #info Proposed Blocker, anaconda, NEW 16:24:28 does this boil down to the same issue that's manifesting in multiple ways or multiple similar issues? 16:24:44 this is another of the infamous unicode bugs, straight up crasher in various informational messages on the INSTALLATION DESTINATION screen if the translation has non-ascii characters. 16:25:06 is this a dupe of 1193389? 16:25:08 tflink: well, it's one of those angels/pinheads things 16:25:12 danofsatx: no. 16:25:55 tflink: all these bugs are cases where a unicode decode/encode operation gets triggered without an encoding being explictly specified 16:26:01 * jreznik suggest to rewrite anaconda in some language that really cares about internationalization... but it's likely not going to be accepted as blocker :D 16:26:25 on the live images the default encoding (sometimes? it's weird, because not *every* case you'd expect to crash does) is ascii, and that causes a crash. 16:26:37 (when the string contains any non-ASCII characters.) 16:27:10 it's actually under active discussion whether we use a kinda ugly hack which should ought to avoid all such issues happening, or go in and fix each case the 'right' way, one by one. 16:27:30 though i suspect the ugly hack's gonna win. at least, i hope so. 16:27:48 'force the default encoding to be utf-8 on lives as well as non-lives' hammer is what I like too 16:28:11 it's probably worth tracking the known crashers separately, though, just in case 16:28:12 at least, both would be the same environments 16:28:19 (and so we can double-check the hammer fixes all of them) 16:28:24 jreznik: yeah, that's principally why i like it. 16:29:18 anyhow, +1 to this for me, criterion probably beta "When using the guided partitioning flow, the installer must be able to: ... Reject or disallow invalid disk and volume configurations without crashing." as this case triggers when you don't select a disk (though various other messages on the same screen would do it too) 16:29:40 we hit it several times, where it worked for developer as he was doing it on non-live and then it was crashing on live and it's hard to make sure all conversions are done properly everwhere 16:29:54 +1 16:30:28 +1 16:31:06 proposed #agreed 1217504 - AcceptedBlocker - violates "When using the guided partitioning flow, the installer must be able to: ... Reject or disallow invalid disk and volume configurations without crashing." in the case of a non-English live install 16:31:17 ack 16:31:45 ack 16:32:03 ack 16:32:14 #agreed 1217504 - AcceptedBlocker - violates "When using the guided partitioning flow, the installer must be able to: ... Reject or disallow invalid disk and volume configurations without crashing." in the case of a non-English live install 16:32:20 #topic (1217610) UnicodeDecodeError when entering encryption passphrase in language other than English on live install 16:32:21 #link https://bugzilla.redhat.com/show_bug.cgi?id=1217610 16:32:21 #info Proposed Blocker, anaconda, NEW 16:32:24 I promise, this is the last of 'em! 16:32:51 another of the unicode bugs, this one happens as soon as you start typing an encryption passphrase if the translation of the 'your passphrase is too short!' message has a non-ASCII character 16:33:32 * jreznik knows why he uses english keyboard everywhere and why he cripples even his name to be ascii :D 16:33:34 I assume that non-ASCII chars are accepted for luks passphrases? 16:33:43 it's not about the character in the passphrase at all 16:33:53 it's a yellow-bar warning message that actually triggers the crash 16:33:55 +1 16:34:22 ah 16:34:24 +1 16:34:33 +1 16:34:35 +1 16:35:15 +1 16:35:48 proposed #agreed 1217610 - AcceptedBlocker - violates "When using the guided partitioning flow, the installer must be able to: ... Complete an installation using any combination of disk configuration options it allows the user to select" for live, non-English installs 16:36:05 secretary note: that's different from the one i proposed it under, so note it. :) 16:36:13 (i think that one fits better) 16:37:00 ack 16:37:02 ack 16:37:08 Does the user spoke also crash when using a translation on the same basis of "Your password is too short!"? 16:37:18 ack 16:37:24 ack 16:38:18 danofsatx: i think that's one of the ones i expected to crash and tested, but it doesn't. there's *some* kind of twist to this bug where the utf-8 encoding actually is being used sometimes, but we haven't been able to pin down when 16:38:31 ok, then...ack 16:38:38 i have a half-assed theory that it's to do with exactly when various imports happen during anaconda's run, but bleh 16:38:46 #agreed 1217610 - AcceptedBlocker - violates "When using the guided partitioning flow, the installer must be able to: ... Complete an installation using any combination of disk configuration options it allows the user to select" for live, non-English installs 16:38:59 #topic (1188282) Wayland: gnome-session does not start ssh-agent 16:38:59 #link https://bugzilla.redhat.com/show_bug.cgi?id=1188282 16:38:59 #info Proposed Blocker, gnome-session, NEW 16:39:13 wayland still isn't blocking stuff, afaik, so -1 16:39:51 -1. 16:39:54 my comment on this bug is crap, btw, i didn't notice that it was wayland 16:40:01 yep, -1 blocker 16:40:55 -1 16:40:57 wait, I thought wayland was the default session. 16:41:05 danofsatx: no. gdm runs on wayland by default, but GNOME does not. 16:41:14 Oh, I get it.... 16:41:28 * danofsatx was breifly confused, but now he's just clueless 16:41:36 proposed #agreed 1188282 - RejectedBlocker - this report is for GNOME on Wayland, which is not the default session and is not considered to block the release in general 16:41:36 -1 16:41:38 -1 16:41:39 ack 16:41:45 ack 16:41:45 ack 16:41:46 ack 16:41:51 ack 16:42:01 #agreed 1188282 - RejectedBlocker - this report is for GNOME on Wayland, which is not the default session and is not considered to block the release in general 16:42:08 #topic (1202943) [abrt] gnome-shell: g_type_check_instance(): gnome-shell killed by SIGSEGV 16:42:08 #link https://bugzilla.redhat.com/show_bug.cgi?id=1202943 16:42:09 #info Proposed Blocker, gnome-shell, NEW 16:43:01 this is the crash that was happening when the site that backs the captive portal URL check was down, iirc 16:43:26 how often it happens? was this first time? 16:43:48 i *think* it might be the case that shell basically crashes any time the captive portal thing pops up 16:43:59 so we all noticed it when the site was down and gnome was wrongly thinking we were all on captive portals 16:44:08 but presumably it'll still crash when you really ARE on a captive portal 16:44:55 even if that's the case, not sure it's blocker material 16:45:20 but it would suck for someone on a livecd to get crashes just because they were on a captive portal 16:45:23 yeah 16:45:31 on the other hand, how often do you use a live image in that situation? 16:45:34 i dunno, seems borderline 16:45:39 definitely +1 FE 16:46:02 if it was just due to a short-term infra problem, -1 blocker 16:46:52 -1/+1 16:46:53 if it does indeed happen for all captive portals ... I'm going to go with -1 as well - same "last blocker before release" test 16:46:59 so -1/+1 16:47:00 * jreznik is inclined to agree with tflink 16:47:10 -1/+1 16:47:23 tflink: luya's comment and blocker proposal indicate he's seeing it in an Actual Real Captive Portal Scenario 16:48:35 I can test this by blocking the URL in my firewall. It would take me a while, though.... 16:48:49 proposed #agreed 1202943 - RejectedBlocker AcceptedFreezeException - if GNOME is crashing whenever it sees a captive portal that's bad, but probably possible to deal with, can be fixed for installed systems with an update, and we suspect lives aren't commonly used in this scenario 16:48:54 if it crashes when you are on a captive portal i would be +1 16:49:00 ack 16:49:04 as we can not fix that with an update for live media 16:49:26 but sure 16:49:26 dgilmore: well, Shell recovers from crashes 16:49:49 iirc when this was happening due to the server bug, it didn't kill your session or anything - i think the actual captive portal window thingy didn't work, but i'm guessing you could just use firefox 16:50:07 we can always go find a captive portal and do some testing, i guess. sounds radical. 16:50:23 I'll do that this afternoon. 16:51:59 #agreed 1202943 - RejectedBlocker AcceptedFreezeException - if GNOME is crashing whenever it sees a captive portal that's bad, but probably possible to deal with, can be fixed for installed systems with an update, and we suspect lives aren't commonly used in this scenario 16:52:07 we can always repropose if it turns out we made a bad call there 16:52:14 #topic (1205725) no visible mouse cursor on some arm boards 16:52:14 #link https://bugzilla.redhat.com/show_bug.cgi?id=1205725 16:52:14 #info Proposed Blocker, xorg-x11-server, VERIFIED 16:52:23 +1 to this, sounds pretty bad for X on ARM 16:52:47 +1 16:52:53 +1 16:53:06 +1 16:53:23 +1 16:53:25 the update fixes things :) 16:54:00 +1 16:54:19 proposed #agreed 1205725 - AcceptedBlocker - this makes X effectively unusable on most ARM platforms, effectively violates all the graphical desktop criteria as it's difficult to do any of them if you can't see the cursor 16:54:35 ack 16:54:36 (fwiw its not really arm specific) 16:54:49 ack 16:54:51 it's not? 16:55:02 * danofsatx hasn't seen it on Intel 16:55:08 its "modesetting driver on device that does not do accel" 16:55:22 happens to be most arm devices but also for instance usb vga "cards" etc. 16:55:25 drago01: which, in practice, boils down to ARM most of the time :) 16:55:26 but mostly visible on arm 16:55:34 ack 16:55:42 ack 16:55:57 but yeah its fixed so meh ;) 16:56:26 ack 16:56:28 #agreed 1205725 - AcceptedBlocker - this makes X effectively unusable on most ARM platforms (and even some non-ARM cases), effectively violates all the graphical desktop criteria as it's difficult to do any of them if you can't see the cursor 16:56:57 re-ack 16:57:10 that wasn't a proposal, it was an on-the-fly, without-a-safety-net change ;) 16:57:16 alrighty, that's all the proposed blockers 16:57:20 let's see what there is to revisit in the accepted 16:57:44 #info moving onto reviewing AcceptedBlockers 16:57:45 #topic (1211746) ValueError: new size is too large 16:57:45 #link https://bugzilla.redhat.com/show_bug.cgi?id=1211746 16:57:45 #info Accepted Blocker, anaconda, ASSIGNED 16:58:21 we reviewed this last week, no movement since then, but dlehman did assign it to himself so i'm guessing that means he'll be working on it 16:58:41 #info dlehman has assigned this to self so he will be working on it; all info appears to be present, this is waiting on developer action 16:59:36 don't think there's much else to say... 16:59:43 #topic (1209941) upgraded system does not reboot automatically, ctrl+alt+del is needed: Failed unmounting /sysroot/proc 16:59:43 #link https://bugzilla.redhat.com/show_bug.cgi?id=1209941 16:59:43 #info Accepted Blocker, fedup-dracut, NEW 17:00:05 we're still waiting for the developer on this one, no response to the ping from last week 17:00:30 8 days to final freeze... 17:00:40 * satellit install does work but not appearant to user 17:00:59 well, i guess we can wait one more week before sending up flares 17:01:17 #info we pinged on this one last week, still no response yet, if we don't hear by next week probably time to go for more drastic action 17:01:29 I'd say we can wait until Thursday before sending up flares, We need to test it. 17:01:50 danofsatx: well, next week is freeze, not go/no-go 17:02:02 go/no-go is 05-21 17:02:03 unnerstoot 17:02:09 the earlier the better though... 17:02:11 #topic (1166650) [abrt] liveusb-creator: connection.py:651:call_blocking:DBusException: org.freedesktop.DBus.Error.UnknownMethod: Method "Get" with signature "ss" on interface "org.freedesktop.DBus.Properties" doesn't exist 17:02:11 #link https://bugzilla.redhat.com/show_bug.cgi?id=1166650 17:02:11 #info Accepted Blocker, liveusb-creator, NEW 17:02:34 kinda the same story, waiting on lmacken here 17:02:40 * satellit see my comments in bug 17:02:46 i guess i'll send out a blocker status review email this week 17:02:53 #info we pinged on this one last week, still no response yet, if we don't hear by next week probably time to go for more drastic action 17:03:42 adamw: and it's time to get my stick ready again for final 17:03:53 *thwack thwack* 17:04:04 #action jreznik to dispense stick-based discipline 17:04:13 #topic (1206760) Plasma desktop doesn't notify for available updates 17:04:13 #link https://bugzilla.redhat.com/show_bug.cgi?id=1206760 17:04:13 #info Accepted Blocker, plasma-pk-updates, ASSIGNED 17:05:09 satellit: if it notifies for updates when booted live that's actually wrong... 17:05:22 * satellit works but is also active as mounted live fixed? yes 17:05:58 only installed systems are supposed to notify, there should be no notifications when booted live 17:06:13 works but slower due to update search 17:09:35 #info in latest nightlies it seems like update notification is now enabled by default, but on lives as well as installed systems; that needs to be fixed 17:09:42 #info KDE team are on it 17:09:55 #topic (1200161) Non-Fatal SELINUX Faults exist during bootup of 22_Alpha_RC3 17:09:56 #link https://bugzilla.redhat.com/show_bug.cgi?id=1200161 17:09:56 #info Accepted Blocker, rng-tools, NEW 17:10:34 we're still awaiting info on this one. IIRC I searched for other reports of the first AVC and found a somewhat confusing chain of dupes 17:11:00 * satellit have to leave..afk 17:11:30 thanks satellit 17:11:55 has anyone else seen "SELinux is preventing rngd from execmod access on the file /usr/sbin/rngd" on recent installs? 17:13:44 * danofsatx hasn't looked 17:14:03 i suggest we give it one more week then drop it if there's no further info from bob 17:14:20 #info still waiting on confirmation from the reporter that this is still a problem, we are not aware of other reports of the same AVC 17:14:24 sounds good 17:14:51 #info if no-one provides info that this is reproducible with current builds by next week, we will drop it 17:15:05 #topic (1190377) SELinux is preventing polkitd from 'read' accesses on the lnk_file localtime. 17:15:05 #link https://bugzilla.redhat.com/show_bug.cgi?id=1190377 17:15:05 #info Accepted Blocker, systemd, NEW 17:15:50 looks like more info is requested from testers here 17:16:22 #info some info is requested in comment #16, we should provide it 17:16:28 does anyone want to volunteer for this one? 17:17:03 I'll spin up a couple VMs. 17:17:09 * danofsatx goes to grab the TC 17:17:10 thanks dan 17:17:48 i'd suggest noting when you start install, running install, noting when it finishes, leaving it 10 mins for clarity, note time again, run through g-i-s, note time when you're done, then wait some time before logging in, or something along those lines 17:18:02 basically, note the times all along, and leave time between significant stages 17:18:35 and remember per https://bugzilla.redhat.com/show_bug.cgi?id=1210045#c5 to reproduce you have to install i386 and create the user with g-i-s apparently 17:18:48 (thoguh i can't see how arch could have anything to do with it) 17:19:02 #action danofsatx to try and provide requested info for this bug 17:20:04 #topic (1200559) [abrt] totem: browse_cb(): totem killed by SIGABRT 17:20:05 #link https://bugzilla.redhat.com/show_bug.cgi?id=1200559 17:20:05 roger, will take that into consideration 17:20:05 #info Accepted Blocker, totem, ASSIGNED 17:20:12 so bastien's been active on this one, but no fix yet 17:20:19 still, don't think any specific action is required 17:23:25 #info sufficient info is provided on this bug, it is just awaiting a fix, developer is active on the bug 17:23:50 alrighty, that's all the non-modified acceptedblockers 17:24:18 we have one proposed FE, but freeze isn't till the day after next week's meeting so we could just wait till then 17:25:03 ....but, we already acked some FEs for today, so what's another 5 mins? 17:25:14 well, sure, then 17:25:30 #info moving onto the sole proposed freeze exception issue 17:25:31 #topic (1218241) CVE-2015-3315 abrt: Various race-conditions and symlink issues found in abrt [fedora-all] 17:25:31 #link https://bugzilla.redhat.com/show_bug.cgi?id=1218241 17:25:31 #info Proposed Freeze Exceptions, abrt, NEW 17:26:06 oh yeah, the fun abrt vuln. 17:27:18 aiui this is a privesc, so it's not super relevant to the live case 17:27:28 +1 FE - it doesn't sound like this would have much impact on live systems and could be fixed with updates if need be 17:27:50 but it's always good to have serious vulns fixed in the release media 17:27:51 as long as the update is submitted early enough 17:27:58 yep, exactly 17:28:00 still, this involves significant change so i might not want it in the last day or two 17:29:52 +1 FE 17:30:10 +1 FE 17:30:38 +1 FE 17:30:43 proposed #agreed 1218241 - AcceptedFreezeException - this doesn't seem like it has a huge impact on any non-updateable use cases, but it is still good to have vulns fixed on the release media to reduce the likelihood of exposure windows between installation and update 17:31:07 * potty salute everyone. 17:31:14 hi potty, we're just about done :) 17:31:21 ack 17:31:40 :-( 17:32:50 heh, flying visit 17:32:53 #agreed 1218241 - AcceptedFreezeException - this doesn't seem like it has a huge impact on any non-updateable use cases, but it is still good to have vulns fixed on the release media to reduce the likelihood of exposure windows between installation and update 17:32:56 alrighty, that's all the bugs 17:33:01 looks like we got through 'em pretty fast 17:33:07 #topic Open floor 17:33:11 any other business, folks? 17:33:31 nothing from me 17:33:51 nope 17:33:56 nada 17:35:36 alrighty, that's what i like to hear :P 17:35:37 On behalf of the fedora-arm user community we would like to thank you all for your efforts in this project. You have enabled those who could not do for themselves without you and have brought a lot of happiness into the world that would not have been without you. We thank you all and look forward to the progress of fedora-arm. 17:35:45 Froolap: well, thanks! 17:35:48 thanks to you too 17:36:18 adamw: thanks for leading 17:36:26 * adamw sets the acme fuse 17:36:29 danofsatx: thanks for secretarializing 17:36:31 it's good to see happines in the world! 17:36:37 and thanks everyone! 17:38:13 okiebye 17:38:54 #endmeeting