15:00:52 #startmeeting Prioritized bugs and issues 15:00:52 Meeting started Wed Sep 26 15:00:52 2018 UTC. 15:00:52 This meeting is logged and archived in a public location. 15:00:52 The chair is bcotton. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:52 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00:52 The meeting name has been set to 'prioritized_bugs_and_issues' 15:00:53 #meetingname Fedora Prioritized bugs and issues 15:00:53 The meeting name has been set to 'fedora_prioritized_bugs_and_issues' 15:01:03 #topic Purpose of this meeting 15:01:05 #info The purpose of this process is to help with processing backlog of bugs and issues found during the development, verification and use of Fedora distribution. 15:01:06 #info The main goal is to raise visibility of bugs and issu 15:01:14 #undo 15:01:14 Removing item from minutes: INFO by bcotton at 15:01:06 : The main goal is to raise visibility of bugs and issu 15:01:26 #info The main goal is to raise visibility of bugs and issues to help contributors focus on the most important issues. 15:01:27 #link https://fedoraproject.org/wiki/Fedora_Program_Management/Prioritized_bugs_and_issues_-_the_process 15:01:34 #topic Roll Call 15:02:21 i'm here 15:02:31 hi, mattdm, i've missed you! 15:02:40 i know it's been so long 15:03:24 let's see if anyone else joins us today. the list is pretty short unless it's gotten longer since yesterday :-) 15:05:06 hokay 15:05:17 #topic Nominated bugs 15:05:18 #link https://bugzilla.redhat.com/buglist.cgi?bug_status=NEW&bug_status=ASSIGNED&bug_status=POST&bug_status=MODIFIED&bug_status=ON_DEV&bug_status=ON_QA&bug_status=VERIFIED&bug_status=RELEASE_PENDING&classification=Fedora&keywords=Triaged&keywords_type=nowords&list_id=9195844&product=Fedora&query_format=advanced&status_whiteboard=PrioritizedBug&status_whiteboard_type=allwords 15:05:20 just the one 15:05:30 #link https://bugzilla.redhat.com/show_bug.cgi?id=1578494 15:05:31 #info 8 other open bugs with the same summary 15:05:39 #info deferred from previous meetings 15:05:41 #link https://retrace.fedoraproject.org/faf/problems/?component_names=PackageKit&associate=__None&type=core&daterange=2018-09-01%3A2018-09-30&bug_filter=None&function_names=&binary_names=&source_file_names=&since_version=&since_release=&to_version=&to_release= 15:06:13 so this one seems sort of related to 1631533, which was rejected as a blocker for F29 final. but everyone agrees it's ugly 15:08:00 we still need a good reproducer 15:08:24 agreed 15:08:53 we've already deferred it twice, so i'm inclined to say we reject it and it can be re-nominated if someone comes up with a good reproducer 15:09:00 maybe now that the beta is out the door we can ask qa to see if they can investigate? 15:09:14 adamw: are you awake yet? 15:09:19 no 15:09:23 dang 15:09:36 what's up? 15:09:53 talking about https://bugzilla.redhat.com/show_bug.cgi?id=1578494 and whether or not it gets the Prioritized stamp 15:10:21 it seems related to the other "i can't run multiple package transations at once" bugs, but we don't seem to have a good reproducer. is this something QA can dig into more? 15:10:38 oh look a package manager crashed? that's unpossible 15:10:47 my inclination is to give this an approval or rejection in this meeting instead of punting on it a third time 15:12:24 i'm having a look at it now 15:15:37 should we go on to other tickets while adam looks? 15:15:38 welp. from a quick look i think this likely always happens outside of an actual package transaction, so it's "just" a crash... 15:15:46 but might be good to get hughsie to confirm that 15:17:05 the other thing to note is that the person who requested PrioritizedBug status still hasn't said what it crashing actually means to him, apart from the fact that it crashes 15:17:24 from the backtrace...huh. repo_internalize_trigger is passed one arg, 'repo', which in this case exists. then it does this: 15:17:29 auto hrepo = static_cast(repo->appdata); 15:17:38 which in this case seems to wind up with hrepo being null. 15:18:11 I mean, core system infrastructure stuff like this should not be crashing as a matter of course 15:18:46 yeah. 15:19:03 there's the question of exactly what the consequences are for e.g. gnome-software and update notifications and things if packagekitd crashes liek this 15:19:14 i don't know offhand the circumstances under which it'd get automatically restarted 15:19:29 but...there are *other* crasher bugs in packagekitd. are we gonna prioritize 'em all? 15:19:39 yeah; if it doesn't get restarted, that's arguably a security bug 15:19:41 my immediate inclination would be -1 or you're risking over-diluting the process. 15:19:49 beacuse people will miss "your system needs updating" notices 15:19:57 i mean, arguably. it's a bit of a stretch. 15:20:49 yeah let's go with "rejected blocker, but we're working with the DNF team to prioritize robustness in general, which this falls under" 15:20:49 i'm also at -1 (without prejudice, as the lawyers would say) 15:20:50 I'd love it if someone could get the libdnf team to debug it, because I've tried to look into this crash and just got lost in libdnf code 15:21:02 there's 5 million dupes of this in bugzilla 15:22:00 proposed #agreed BZ 1578494 is rejected as a PrioritizedBug. It can be re-nominated with a reproducer or if the impact is more severe 15:22:01 kalev: dupes of that one, or things that look similar? 15:22:35 mattdm: lots of bugs where packagekit crashes in repo_internalize_trigger 15:23:53 kalev: i mean, the dumb trivial solution is "have repo_internalize_trigger check whether hrepo exists after the cast", no? 15:24:03 you could send a pr that does that, and see if it triggers them into finding the real fix. :P 15:24:18 or i could...heh 15:25:10 sure :) 15:25:41 okay, well no one has complained about my wording, so i'mma stamp it 15:25:43 mattdm: https://www.google.com/search?q=repo_internalize_trigger+site%3Abugzilla.redhat.com&ie=utf-8&oe=utf-8&client=firefox-b-ab 15:25:48 it seems quite likely those are all the same 15:25:52 or at least lots of them 15:26:27 * adamw will do a dupe run 15:26:33 #agreed BZ 1578494 is rejected as a PrioritizedBug. It can be re-nominated with a reproducer or if the impact is more severe 15:26:40 good news. that was the only new one 15:26:54 so we'll do a quick flyby of the existing prioritized bugs and maybe even finish ahead of schedule 15:27:13 #topic Accepted bugs 15:27:15 #link https://bugzilla.redhat.com/buglist.cgi?bug_status=NEW&bug_status=ASSIGNED&bug_status=POST&bug_status=MODIFIED&bug_status=ON_DEV&bug_status=ON_QA&bug_status=VERIFIED&bug_status=RELEASE_PENDING&classification=Fedora&keywords=Triaged&keywords_type=allwords&list_id=9195442&product=Fedora&query_format=advanced&status_whiteboard=PrioritizedBug&status_whiteboard_type=allwords 15:27:25 #link https://bugzilla.redhat.com/show_bug.cgi?id=1336435 15:27:26 #info BZ 13364535 was accepted on 2017-08-02 15:27:32 #info assignee has made no comments since needinfo flag was set on 2017-09-27 15:27:33 #info Upstream bug has had no activity in the last year 15:27:37 #link https://gitlab.gnome.org/GNOME/gnome-software/issues/116 15:27:50 mattdm, you were going to follow up with Richard to update this? 15:28:25 yes but i haven't. 15:28:38 can i have one more meeting, please? :) 15:28:42 #action mattdm to follow up with Richard Hughes to update this bug 15:28:47 only because i like you :-) 15:28:58 next! 15:29:01 #link https://bugzilla.redhat.com/show_bug.cgi?id=1385432 15:29:02 #info BZ 1385432 was accepted on 2017-05-25 15:29:05 #info assignee has made no comments since needinfo flag was set on 2018-08-02 15:29:20 mattdm: you were going to escalate this one internally. i think i saw that email? 15:30:29 * mattdm waits for bug to load 15:30:57 yeah. i guess i need to follow up on that again because not sure anything happened 15:31:23 i read this one yesterday and if anything i'm more confused than ever. it doesn't seem like anyone agrees where the problem is 15:31:34 but i'll #action you to follow up on your escalation 15:31:43 yeah. doing that right now. 15:31:52 #action mattdm to follow up on his internal escalation of this issue 15:32:03 yeah, the problem is that every team seems to point to the other one, and no one will take ownership 15:32:06 yet the problem persists. 15:32:26 you'd think if the buck gets passed often enough, it would just disappear :-) 15:32:42 #link https://bugzilla.redhat.com/show_bug.cgi?id=1306992 15:32:43 #info BZ 1306992 was accepted on 2018-08-29 15:33:27 no movement on this one, but i think we can give it one more meeting before we start pushing. it's at least partially solved at this point 15:34:19 yeah there was some discussion on solutions so let's give it a little time 15:35:12 #link https://bugzilla.redhat.com/show_bug.cgi?id=1548586 15:35:13 #info BZ 1548586 was accepted on 2018-06-11 15:35:15 #info fixed version in dnf 3.0 COPR, but not available in Fedora yet 15:35:33 #info We want the fix backported to F28, which does not appear to have happened 15:36:02 i can follow up with daniel about fixing it in f28 15:36:46 yes please. thanks 15:37:13 #action bcotton to follow up with Daniel Mach on fixing BZ 154856 in F28 15:37:27 #link https://bugzilla.redhat.com/show_bug.cgi?id=1558485 15:37:28 #info BZ 1558485 was accepted on 2018-08-02 15:37:29 #info Fedora overrides this with a config setting 15:37:48 since we override this with a config, i think we can call this resolved for the purposes of PrioritizedBug status 15:39:24 +1 15:40:40 #agreed BZ 1558485 is considered resolved for the purpoes of PrioritizedBug status, since Fedora ships with a config that overrides the upstream issue 15:40:54 and that, my friends, is the end of the list 15:41:39 thank you bcotton! 15:41:46 #topic Next meeting 15:41:47 #info The next meeting is 10 October at 1500 UTC in #fedora-meeting 15:41:55 see you then! 15:42:19 anything else before i bang the gavel? 15:43:04 too late! 15:43:06 #endmeeting