14:00:35 <bcotton> #startmeeting Prioritized bugs and issues
14:00:36 <zodbot> Meeting started Wed Aug 29 14:00:35 2018 UTC.
14:00:36 <zodbot> This meeting is logged and archived in a public location.
14:00:36 <zodbot> The chair is bcotton. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:36 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
14:00:36 <zodbot> The meeting name has been set to 'prioritized_bugs_and_issues'
14:00:45 <bcotton> #meetingname Fedora Prioritized bugs and issues
14:00:46 <zodbot> The meeting name has been set to 'fedora_prioritized_bugs_and_issues'
14:00:54 <bcotton> #topic Purpose of this meeting
14:01:05 <bcotton> #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.
14:01:07 <bcotton> #info The main goal is to raise visibility of bugs and issues to help  contributors focus on the most important issues.
14:01:13 <bcotton> #link https://fedoraproject.org/wiki/Fedora_Program_Management/Prioritized_bugs_and_issues_-_the_process
14:01:20 <bcotton> #topic Roll Call
14:01:36 * jsmith is here, but lurking (due to work meetings)
14:02:57 <bcotton> hello, jsmith
14:03:45 * zbyszek is here just in case
14:03:52 <bcotton> we'll give folks another moment to wander in
14:03:59 <bcotton> hello, zbyszek
14:04:04 <zbyszek> hi
14:04:07 <mattdm> hello!
14:04:19 <mattdm> sorry, i dunno what's up with this -2 business :)
14:04:36 <bcotton> don't worry, it'll move for subsequent meetings (spoiler alert)
14:05:30 <mattdm> heh
14:05:33 <bcotton> okay, let's get started
14:05:39 <bcotton> #topic Nominated bugs
14:05:48 <bcotton> #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
14:05:59 <bcotton> #link https://bugzilla.redhat.com/show_bug.cgi?id=1026119
14:06:00 <bcotton> #info deferred from previous meeting
14:06:01 <bcotton> #info reporter says behavior has gone away on some F28 systems and persists on others.
14:06:41 <mattdm> zbyszek: thanks for the update here
14:06:56 <bcotton> zbyszek++
14:06:56 <zodbot> bcotton: Karma for zbyszek changed to 12 (for the current release cycle):  https://badges.fedoraproject.org/tags/cookie/any
14:07:03 <mattdm> zbyszek++
14:07:03 <zodbot> mattdm: Karma for zbyszek changed to 13 (for the current release cycle):  https://badges.fedoraproject.org/tags/cookie/any
14:07:17 <bcotton> zbyszek is raking in the karma today :-)
14:07:30 <zbyszek> That's what I'm here for ;)
14:07:40 <bcotton> i'm inclined to vote reject on this one
14:07:41 <mattdm> zbyszek: soooo, is there an upstream fix for this, or something someone in Fedora can work on?
14:08:17 <zbyszek> No, no fix. It's something that could be written, but would require working on the core of pid1, to make the dependency scheme a bit different.
14:08:25 <zbyszek> So an interesting project, but non-trivial.
14:08:37 <zbyszek> At least that's my understanding of the issue.
14:09:34 <mattdm> Okay so, proposal coming give me a sec :)
14:10:27 <mattdm> #proposal The effects of this are just cosmetic, but it's still a real bug. Unfortunately, the fix is non-trivial. We recommend closing this as upstream (and making sure an upstream bug is filed so the issue can eventually be addressed).
14:12:05 <bcotton> +1
14:12:24 <jsmith> +1 from me
14:12:31 <mattdm> (I'm +1 to my own proposal fwiw)
14:12:35 <bcotton> zbyszek?
14:12:49 <zbyszek> works for me
14:13:02 <mattdm> The followup question is: *is* there an existing upstream bug report?
14:13:34 <bcotton> #agreed The effects of BZ 1026119 are just cosmetic, but it's still a real bug. Unfortunately, the fix is non-trivial. We recommend closing this as upstream (and making sure an upstream bug is filed so the issue can eventually be addressed).
14:13:51 <zbyszek> I don't think so. So probably this Fedora bug shouuldn't be closed, unless somebody wants to spend time copying the comments and all info.
14:14:15 <bcotton> or just link to the RHBZ from upstream?
14:14:49 <zbyszek> There isn't a big split between upstream and Fedora, so it's better to just keep the Fedora bug open.
14:14:55 <bcotton> ok
14:15:14 <bcotton> but we all agree that it's not a PrioritizedBug
14:15:39 <mattdm> yah, and I guess we should revise the proposal if zbyszek doesn't want it closed
14:15:47 <bcotton> #undo
14:15:47 <zodbot> Removing item from minutes: AGREED by bcotton at 14:13:34 : The effects of BZ 1026119 are just cosmetic, but it's still a real bug. Unfortunately, the fix is non-trivial. We recommend closing this as upstream (and making sure an upstream bug is filed so the issue can eventually be addressed).
14:16:47 <bcotton> #propsal  The effects of BZ 1026119 are just cosmetic, but it's still a real bug. Unfortunately, the fix is non-trivial. We reject it as a PrioritizedBug and suggest the reporter file an issue upstream
14:16:47 <mattdm> last sentence to "systemd developers recommend leaving bug open so it can be tracked and eventually addressed, but we're removing this from the prioritized bug list."?
14:16:49 <bcotton> ?
14:17:19 <mattdm> If upstream isn't really making a strong distinction and is following fedora bugs in the rh bugzilla, no point in making them refile
14:17:45 <bcotton> #undo
14:17:45 <zodbot> Removing item from minutes: INFO by bcotton at 14:06:01 : reporter says behavior has gone away on some F28 systems and persists on others.
14:17:52 <bcotton> okay, let's give this one last try
14:18:14 <bcotton> #proposal  The effects of BZ 1026119 are just cosmetic, but it's still a real bug. Unfortunately, the fix is non-trivial. systemd developers recommend leaving bug open so it can be tracked and eventually addressed, but we're removing this from the prioritized bug list.
14:18:25 <bcotton> (this time with "proposal" spelled correctly)
14:18:55 <mattdm> +1
14:19:26 <zbyszek> looks good
14:20:11 <bcotton> #agreed The effects of BZ 1026119 are just cosmetic, but it's still a real bug. Unfortunately, the fix is non-trivial. systemd developers recommend leaving bug open so it can be tracked and eventually addressed, but we're removing this from the prioritized bug list.
14:20:23 <bcotton> okay, next!
14:20:27 <bcotton> #link https://bugzilla.redhat.com/show_bug.cgi?id=1575281
14:20:29 <bcotton> #info deferred from previous meeting to get input from Workstation WG
14:20:30 <bcotton> #info Workstation WG is weakly in favor of accepting
14:20:47 <mattdm> What does "weakly" mean?
14:21:29 <mattdm> I mean, prioritized bugs can't _really_ be weakly accepted. Either someone is gonna be committed to working on it, or not
14:21:42 <bcotton> #link https://lists.fedoraproject.org/archives/list/desktop@lists.fedoraproject.org/message/UUW6P4BYFZABAK5ZBFQD3MUNMHB7ZGYC/
14:22:18 <bcotton> summary: they (read: michael) made an argument for including it, but didn't sound enthusiastic about it
14:23:44 <mattdm> I'm not experiencing this.
14:24:00 <mattdm> (although I am on xorg for $reasons)
14:24:28 <zbyszek> FWIW, I *am* experiencing something similar, but just a second or two, which is noticable, but not unpleasant enough to really matter
14:25:12 <bcotton> i use KDE so i can't speak to it
14:25:26 <mattdm> I feel like this belongs on a "desktop polish" list rather than the prioritized bugs list
14:25:40 <mattdm> and I feel like the desktop team should keep that list, not us :)
14:25:53 <bcotton> excellent delegation :-)
14:26:32 <mattdm> We tend to have a lot of desktop issues on the prioritized lists, and I'd like to reserve that for things where I feel good about pulling strings and asking for shifted priorities
14:26:52 <bcotton> #proposal BZ 1575281 is rejected as a PrioritizedBug
14:26:57 <mattdm> and while I don't dispute that polish like this is important, I have a hard time asking to have people pulled from other work for it
14:27:30 <mattdm> I'll ask the Workstation WG about keeping a "desktop polish" list
14:29:34 <bcotton> mattdm, zbyszek, jsmith: any objection to rejecting BZ 1575281?
14:30:14 <zbyszek> I don't have an opinion, I don't know what resources are available
14:30:22 <mattdm> oh sorry I'm +1 to that
14:30:30 <mattdm> +1 to rejecting :)
14:30:57 <bcotton> sounds good enough to me
14:31:01 <bcotton> #agreed BZ 1575281 is rejected as a PrioritizedBug
14:31:17 <bcotton> #action mattdm to ask the Workstation WG about keeping a "desktop polish" list
14:31:37 <bcotton> #link https://bugzilla.redhat.com/show_bug.cgi?id=1306992
14:31:38 <bcotton> #info re-opened bug previously marked fixed
14:33:07 <zbyszek> FWIW, I think a tmpfiles.d path should *always* be included for caches
14:33:27 <zbyszek> Essentially, even if the program in question cleans up properly, it can only do cleanup if is is running.
14:33:49 <mattdm> zbyszek: that's a good idea
14:34:07 <zbyszek> So for the case where the user uninstalls or disables a program, and also for the case where the cleanup malfunctions for whatever reason, there should be delayed tmpfiles.d cleanup scheduled.
14:34:15 <zbyszek> Belt and suspenders.
14:34:25 <zbyszek> The same thing came up with dnf in the past.
14:34:31 <zbyszek> And with yum before that.
14:35:36 <bcotton> so would that be a change to the packaging guidelines?
14:35:41 <mattdm> Ah, from bodhi:
14:35:43 <mattdm> "
14:35:45 <mattdm> PackageKit 1.1.9 adds basic cache cleanup, removing old cache directories for previous release versions (24, 25, 26 etc) when running on Fedora 27 -- something that people have been asking for a while. (There's more cache cleanup fixes to follow in the future, this is just a first step.)
14:35:47 <mattdm> "
14:36:50 <zbyszek> bcotton: I wouldn't put this in the guidelines, because what needs to be done is different for every daemon, and there's just a few of them.
14:37:00 <bcotton> zbyszek: okay makes sense
14:37:58 <bcotton> so if i understand the current state of this bug, it's not relevant anymore if old releases were what's filling up the cache, but it's still possible to hit it with the cache from the current release
14:38:10 <mattdm> yeah, that's my understanding too
14:38:43 <mattdm> zbyszek: in thinking about it more, I'm worried about a too-aggressive tmpfiles.d approach causing a lot of duplicate downloads
14:38:46 <bcotton> having PackageKit executing a slow-burn DoS against users hard drives seems bad
14:39:40 <zbyszek> mattdm: I'd put it at 1 month or so. If nothing accesses the file within one month, delete it.
14:41:15 <bcotton> so how do we want to proceed with this particular bug?
14:41:40 <mattdm> bcotton I left a note in the bug. I'll also bring the tmpwatch suggestion to hughsie
14:41:46 <mattdm> i mean tmpfiles.d
14:41:58 <zbyszek> It sounds like the fix currently in packagekit is enough to reject it as prioritized
14:42:27 <mattdm> zbyszek: I don't think so. the concern I have is machines that get rebooted for updates on, say, a monthly schedule
14:42:38 <mattdm> these will accumulate a lot of cruft
14:42:49 <mattdm> maybe the current fix *plus* tmpfiles.d
14:44:11 <bcotton> #proposal BZ 1306992 is rejected as a PrioritzedBug but can be re-nominated in the future if subsequent PackageKit releases do not address the "current-release fills the filesystem" scenario
14:44:49 <mattdm> I'm -0.5 ... I'd like to keep it on the radar
14:45:52 <bcotton> does that mean accepting or deferring for 2 weeks?
14:46:16 <mattdm> did it get unaccepted? If so it means putting it back to accepted
14:46:34 <bcotton> it's currently neither accepted nor rejected
14:46:56 <mattdm> okay, so, I guess I'm for putting it to accepted
14:47:08 <bcotton> okay
14:47:27 <bcotton> #proposal BZ 1306992 is accepted as a PrioritizedBug
14:47:32 <mattdm> +1
14:47:32 <bcotton> i am +1 to accepting
14:47:39 <bcotton> zbyszek, jsmith?
14:47:53 <zbyszek> +1
14:47:55 <jsmith> +1 to the proposal
14:48:05 <bcotton> #agreed BZ 1306992 is accepted as a PrioritizedBug
14:48:20 <bcotton> okay, that's all the candidate bugs
14:48:31 <bcotton> but we have a few bugs that have been on the list for a while with no movement
14:48:35 <jsmith> (Sorry for the latency, juggling two things and work plus IRC meetings)
14:49:16 <bcotton> #topic Accepted bugs
14:49:17 <bcotton> #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
14:49:25 <bcotton> #link https://bugzilla.redhat.com/show_bug.cgi?id=1556790
14:49:27 <bcotton> #info BZ 1556790 was accepted on 2018-04-25
14:49:28 <bcotton> #info assignee has made no comments and needinfo flag remains set from 2018-08-02
14:49:35 <bcotton> #link https://bugzilla.redhat.com/show_bug.cgi?id=1336435
14:49:37 <bcotton> #info BZ 13364535 was accepted on 2017-08-02
14:49:38 <bcotton> #info assignee has made no comments since needinfo flag was set on 2017-09-27
14:49:43 <bcotton> #link https://bugzilla.redhat.com/show_bug.cgi?id=1385432
14:49:44 <bcotton> #info BZ 1385432 was accepted on 2017-05-25
14:49:46 <bcotton> #info assignee has made no comments since needinfo flag was set on 2018-08-02
14:49:54 <bcotton> #link https://bugzilla.redhat.com/show_bug.cgi?id=1563674
14:49:56 <bcotton> #info BZ 1563674 was accepted on 2018-08-01
14:49:57 <bcotton> #info assignee has made no comments since needinfo flag was set on 2018-06-06
14:50:22 <bcotton> so here's where our process is weak: what do we do with PrioritizedBugs that don't get actual priority?
14:51:09 <mattdm> bcotton: you and I, or relevant WG members, are supposed to round up help.
14:52:01 <mattdm> FWIW I think #1556790 should be removed because the WG removed the auto-suspend behavior
14:52:50 <bcotton> mattdm: have a reference link handy for the auto-suspend removal?
14:53:45 <mattdm> bcotton: not handy no :(
14:54:05 <mattdm> mclasen: do you have a reference to the plans for auto-suspend at hand?
14:54:11 <mattdm> topic is https://bugzilla.redhat.com/show_bug.cgi?id=1556790
14:54:26 * mclasen should stop idling in meeting channels
14:54:27 <mattdm> which I recommend removing from prioritized list if we're not auto-suspending by default
14:54:42 <mattdm> mclasen: I thought you were doing it on purpose to be helpful
14:55:09 <mattdm> It's incredibly useful to have someone from desktop team available for this meeting
14:55:15 <mattdm> even if usually the whole thing isn't of interest
14:55:24 <mclasen> this is an f28 bug...
14:55:27 <mattdm> is there someone else who would be better?
14:55:45 <mattdm> mclasen: yes, f28 is the current release
14:55:58 <mclasen> I don't even know what this meeting is, and what prioritized bugs are, so my usefulness is offset by my cluelessness
14:56:37 <mclasen> in any case, there's no active plans around automatic suspend that I am aware of
14:56:51 <mattdm> This is a process for getting attention to important *non blocker* bugs
14:56:56 <mclasen> we tried it for f28, and were forced to revert. I don't think it has been revisited
14:57:06 <mattdm> mclasen: do you have a link to the revert decision?
14:57:14 <bcotton> so i see this https://pagure.io/fedora-workstation/issue/42#comment-504425
14:57:18 <mclasen> getting attention ... of people idling in this channel ?
14:57:46 <mattdm> mclasen: getting attention of people who can address the problems we identify as high impact
14:58:53 <bcotton> #proposal Remove BZ 1556790 from PrioritzedBugs since auto-suspend was removed by Workstation WG per https://pagure.io/fedora-workstation/issue/42#comment-504425
14:58:58 <mattdm> +1
14:59:03 <bcotton> +1
14:59:06 <mattdm> I have another meeting I need to run to
14:59:14 <mattdm> bcotton: we can talk more about process here later
14:59:26 <zbyszek> It looks like https://gitlab.gnome.org/GNOME/mutter/merge_requests/54 would solve the bug properly
14:59:38 <bcotton> mattdm: sounds good. in the meantime, i'll start pestering non-responsive assignees out-of-band to see if that gets a reaction
14:59:47 <zbyszek> It was merged 5 months ago, not sure in what release it ended up.
14:59:47 <bcotton> #action bcotton to followup with non-responsive assignees
14:59:59 <zbyszek> +1 to the proposal
15:00:34 <bcotton> #agreed Remove BZ 1556790 from PrioritzedBugs since auto-suspend was removed by Workstation WG per https://pagure.io/fedora-workstation/issue/42#comment-504425
15:00:42 <bcotton> #topic Next meeting
15:00:43 <bcotton> #info Meeting time has been changed to 1500 UTC in #fedora-meeting
15:00:58 <bcotton> okay, that's all the time we have. thanks everyone!
15:01:03 <bcotton> #endmeeting