15:00:14 <sgallagh> #startmeeting FESCO (2018-04-06)
15:00:14 <zodbot> Meeting started Fri Apr  6 15:00:14 2018 UTC.  The chair is sgallagh. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:14 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:00:14 <zodbot> The meeting name has been set to 'fesco_(2018-04-06)'
15:00:14 <sgallagh> #meetingname fesco
15:00:14 <sgallagh> #chair maxamillion dgilmore nirik jsmith sgallagh bowlofeggs tyll jwb zbyszek
15:00:14 <sgallagh> #topic init process
15:00:14 <zodbot> The meeting name has been set to 'fesco'
15:00:14 <zodbot> Current chairs: bowlofeggs dgilmore jsmith jwb maxamillion nirik sgallagh tyll zbyszek
15:00:21 <zbyszek> .hello2
15:00:21 <nirik> morning
15:00:22 <zodbot> zbyszek: zbyszek 'Zbigniew Jędrzejewski-Szmek' <zbyszek@in.waw.pl>
15:00:26 <sgallagh> .hello2
15:00:27 <zodbot> sgallagh: sgallagh 'Stephen Gallagher' <sgallagh@redhat.com>
15:00:31 <zbyszek> good afternoon
15:00:42 <maxamillion> .hello2
15:00:43 <sgallagh> I'm not sure we will have quorum today; a lot of people are out
15:00:43 <zodbot> maxamillion: maxamillion 'Adam Miller' <maxamillion@gmail.com>
15:00:54 <zbyszek> we need one more
15:01:08 <maxamillion> is there a conference or something or is it just bad luck?
15:01:09 <sgallagh> zbyszek: two, I think. You're counting my +1s from the ticket too :)
15:01:19 <sgallagh> No idea
15:01:30 <zbyszek> we have sgallagh, nirik, zbyszek, maxamillion = 4
15:01:59 <sgallagh> ah
15:02:06 <nirik> theres a number of tickets with people voting in them... so that + what we have could solve those.
15:02:16 <sgallagh> yeah, was just typing the same
15:03:38 <sgallagh> I've had meetings all morning, so my script isn't well-prepared today; give me a moment to organize please
15:04:52 <sgallagh> #info Since we don't have full quorum, we'll process the tickets that have votes in them today
15:05:16 <dperpeet> .hello2
15:05:17 <zodbot> dperpeet: dperpeet 'None' <dperpeet@redhat.com>
15:06:09 <sgallagh> #topic #1868 Change the Changes template
15:06:10 <sgallagh> .fesco 1868
15:06:12 <zodbot> sgallagh: Issue #1868: Change the Changes template - fesco - Pagure - https://pagure.io/fesco/issue/1868
15:07:00 <sgallagh> (oops, skipped a ticket I didn't mean to; will go back after this)
15:07:26 <zbyszek> I think we should ask the OP to provide a better template proposal.
15:07:37 <zbyszek> There have been comments, but so far no response to them.
15:07:41 * sgallagh nods
15:07:47 <maxamillion> +1
15:08:04 <sgallagh> zbyszek: Do you want to count that as a -1 on the current proposal and kick this back for improvement?
15:08:17 <zbyszek> Yes.
15:08:35 <sgallagh> OK, with that we cannot have a quorum vote today.
15:09:22 <zbyszek> Actually I wouldn't read jsmith +1 as voting for the current proposal. His comments makes that clear.
15:09:33 <sgallagh> #info With disagreement in the attending members, no binding vote will be possible today. Request is for the proposal to be updated with a better template.
15:10:01 <sgallagh> #topic #1867 F29 System Wide Change: Python 3.7
15:10:01 <sgallagh> .fesco 1867
15:10:03 <zodbot> sgallagh: Issue #1867: F29 System Wide Change: Python 3.7 - fesco - Pagure - https://pagure.io/fesco/issue/1867
15:10:15 <zbyszek> many votes in the ticket...
15:10:55 <nirik> +1 here
15:11:11 <sgallagh> Oh, I missed zbyszek's vote. This is technically alreadyh approved.
15:11:43 <sgallagh> #agreed F29 System Wide Change: Python 3.7 is approved (+6, 0, -0)
15:11:53 <sgallagh> #topic #1869 F29 System Wide Change: Ruby on Rails 5.2
15:11:54 <sgallagh> .fesco 1869
15:11:55 <zodbot> sgallagh: Issue #1869: F29 System Wide Change: Ruby on Rails 5.2 - fesco - Pagure - https://pagure.io/fesco/issue/1869
15:12:13 <sgallagh> Same here
15:12:28 <sgallagh> #agreed F29 System Wide Change: Ruby on Rails 5.2 is approved (+6, 0, -0)
15:12:36 <sgallagh> #topic #1870 F28 Self Contained Change: Stop building 389-ds-base on i686
15:12:36 <sgallagh> .fesco 1870
15:12:36 <zbyszek> ack
15:12:37 <zodbot> sgallagh: Issue #1870: F28 Self Contained Change: Stop building 389-ds-base on i686 - fesco - Pagure - https://pagure.io/fesco/issue/1870
15:13:27 <nirik> still some unknowns around this IMHO, but I'm ok approving the change for now...
15:13:31 <sgallagh> This is looking for one more +1
15:13:40 <sgallagh> And there it is.
15:13:50 <zbyszek> There's 6 in the ticket
15:14:06 <sgallagh> oh, I miscounted somehow. Badly
15:14:06 <zbyszek> And one from nirik here.
15:14:15 <nirik> sure.
15:14:29 <sgallagh> #agreed F28 Self Contained Change: Stop building 389-ds-base on i686 is approved (+7, 0, -0)
15:14:40 <sgallagh> #topic #1871 F29 Self Contained Change: Ansible python3 default
15:14:40 <sgallagh> .fesco 1871
15:14:42 <zodbot> sgallagh: Issue #1871: F29 Self Contained Change: Ansible python3 default - fesco - Pagure - https://pagure.io/fesco/issue/1871
15:15:08 <zbyszek> Yeah, this outcome is not perfect, but at least the change is clearly communicated (re previous topic)
15:15:39 <sgallagh> This one also hit the threshold a few minutes ago
15:15:53 <sgallagh> #agreed F29 Self Contained Change: Ansible python3 default is approved (+5, 0, -0)
15:15:55 <nirik> Not sure what else we can do here... it may be rocky for a bit as people switch
15:16:20 <maxamillion> +1
15:16:27 <maxamillion> sorry
15:16:28 <sgallagh> #undo
15:16:28 <zodbot> Removing item from minutes: AGREED by sgallagh at 15:15:53 : F29 Self Contained Change: Ansible python3 default is approved (+5, 0, -0)
15:16:28 <maxamillion> bleh
15:16:30 <maxamillion> I'm failing at this
15:16:34 <sgallagh> #agreed F29 Self Contained Change: Ansible python3 default is approved (+6, 0, -0)
15:16:40 <sgallagh> It's all good
15:16:45 <zbyszek> It seems that the early dropping of python2- packages is already causing some problems...
15:17:02 <nirik> yep. :(
15:17:08 <sgallagh> It's going to cause problems whenever it happens
15:17:11 <nirik> broke rawhide yesterday too sadly.
15:17:15 <sgallagh> Might as well prepare people for it
15:18:24 <sgallagh> Anything else we want to say here?
15:19:10 <sgallagh> #topic #1872 Disable Test Gating requirements until more UI is enabled
15:19:10 <sgallagh> .fesco 1872
15:19:12 <zodbot> sgallagh: Issue #1872: Disable Test Gating requirements until more UI is enabled - fesco - Pagure - https://pagure.io/fesco/issue/1872
15:19:17 <dperpeet> I would like to point out that I don't think this ticket is clear enough and might conflate issues... also, it would be good to have bowlofeggs here for it, since it mostly concerns bodhi
15:19:20 <maxamillion> +1
15:19:33 <sgallagh> I'd like to propose that we defer this discussion for a week until bowlofeggs can attend
15:19:40 <sgallagh> As he's heavily involved in the process
15:19:49 <nirik> yeah. additionally, there's a infrastructure hackfest next week, and we might be able to knock out some improvements here
15:19:55 <nirik> sgallagh: +1
15:20:21 <zbyszek> ack, we can't make any binding decisions anyway today
15:20:36 <dperpeet> thank you
15:21:18 <sgallagh> #info FESCo would prefer to discuss this when Randy Barlow is present. Also there is a hackfest next week that will aim to improve the situation.
15:21:21 <sgallagh> Good?
15:21:35 <zbyszek> Yes.
15:21:45 <dperpeet> yay
15:22:03 <nirik> ack
15:22:43 <sgallagh> #topic #1873 GNOME suspend to RAM by default
15:22:43 <sgallagh> .fesco 1873
15:22:45 <zodbot> sgallagh: Issue #1873: GNOME suspend to RAM by default - fesco - Pagure - https://pagure.io/fesco/issue/1873
15:23:15 * nirik is in favor of disabling it on live media at the least. Past that not sure...
15:23:37 <sgallagh> I'd like to state first that I *do* think this is mostly up to the Workstation WG to decide, BUT
15:23:43 <nirik> that also
15:24:08 * Southern_Gentlem hopes that WS group follows nirik suggestion
15:24:10 <sgallagh> I think FESCo should rule that their ultimate implementation must not implicitly cause this behavior on non-Workstation installs
15:24:26 <sgallagh> Thereby giving them their autonomy
15:24:32 <zbyszek> The workstation group can decide for workstation, but then it is necessary to ensure that just installing and starting gdm does not have the same effect on non-workstation.
15:24:34 <maxamillion> yeah, I would think that's a Workstation WG decision
15:24:43 <nirik> kalev's suggestion might work... off default, workstation overrides to enable.
15:24:58 <sgallagh> The exact implementation isn't important.
15:25:02 <nirik> right.
15:25:22 <sgallagh> But I think it's WS's responsibility to create the hooks for configuration and let us know what they are
15:25:36 <nirik> this was also filed only 2 days ago... so there's still discussion and ideas
15:25:40 <zbyszek> That said, I use workstation on my workstation, and I don't find the autosuspend default very appealing.
15:25:42 <sgallagh> Right now, my best understanding is that changing the default requires an undocumented gsettings command
15:26:00 <nirik> you can do it from the control-center...
15:26:03 <nirik> (I think)
15:26:10 <sgallagh> nirik: Not for GDM
15:26:15 <sgallagh> Which is the part that breaks Servr
15:26:19 <sgallagh> *Server
15:26:23 <nirik> right, thats true.
15:26:42 * sgallagh now wonders if there's a dating site for EngOps... Servr
15:27:10 <nirik> I am not sure how many people run gdm on servers...
15:27:26 * satellit https://gitlab.gnome.org/GNOME/gnome-settings-daemon/commit/2fdb48fa3333638cee889b8bb80dc1d2b65aaa4a
15:27:29 <sgallagh> nirik: Far more than I would prefer, unfortunately
15:27:51 <sgallagh> There are plenty of people who stick the GUI env on it for convenience and some out of necessity
15:28:08 <sgallagh> (There are plenty of applications out there that have a graphical installer because they ported it from Windows)
15:28:13 <zbyszek> nirik: in my experience it's common to have workstations that double as computation nodes at universities
15:28:27 <nirik> so, how about we ask that the WS WG comes up with some way to not enable this on server installs and let them decide the rest?
15:28:47 <nirik> zbyszek: sure, but I would think they would disable this in their installs then...
15:28:52 <sgallagh> nirik: I think you just restated my proposal, so +1 :)
15:28:59 <nirik> sgallagh: yes. :)
15:29:12 <nirik> but I guess we don't have quorum, so ?
15:29:36 <sgallagh> nirik: We can probably call it a non-binding referrendum to provide feedback at least
15:29:54 <nirik> sure.
15:29:54 <mclasen> sgallagh: gsettings is fully documented, afaik
15:30:06 * nirik notes he has a hard stop at the top of the hour today.
15:30:16 <sgallagh> Something like "FESCo members agree that this would not be a FESCo matter to decide IF ..."
15:30:40 <sgallagh> mclasen: Welcome!
15:31:12 <sgallagh> If you read the scrollback, we're in general agreement that this is Workstation's business, as long as we can ensure that non-WS installs that end up with GDM don't hit it.
15:31:19 <zbyszek> It's hard for me to gather enthusiasm for a +1. I appreciate WG independence, but in this case they seems to be going down the path of least resistance, i.e. just follow the upstream default, without regard how well that will be received.
15:31:21 <sgallagh> Is that something you can work with?
15:31:33 <zbyszek> sgallagh: I wouldn't agree
15:31:49 <sgallagh> zbyszek: Take that to their WG then. We delegated their autonomy previously.
15:31:55 <mclasen> sure, I can
15:32:05 <sgallagh> Taking it away without a REALLY compelling cause seems unfair
15:32:58 <zbyszek> Yes, I understand that. But saying that there's "a general agreement" based on two voices is premature, imho.
15:33:22 <sgallagh> zbyszek: OK, that's fair.
15:33:28 <sgallagh> maxamillion: Want to weigh in? :)
15:33:38 <zbyszek> Three actually, I guess maxamillion is also OK.
15:34:15 <sgallagh> zbyszek: OK, so are you taking a strong stand against this recommendation?
15:34:16 <maxamillion> sorry
15:34:21 <maxamillion> yeah
15:34:22 <maxamillion> +1
15:34:28 <sgallagh> (obviously, we don't have quorum for a binding decision today)
15:34:29 <nirik> well, given that this was filed 2 days ago we could just punt for more discussion...
15:34:29 <maxamillion> I'm multi-tasking horridly this morning
15:35:07 <sgallagh> nirik: Sure, with a wink and a nudge that solving the Server problem would likely result in us kicking the issue back to Workstation
15:35:44 <sgallagh> I won't put that in the #info, FTR
15:35:59 <nirik> well, from some of us anyhow... but sure.
15:36:05 <sgallagh> ok
15:36:24 <mclasen> can I ask though: does this go for any packages that are not in the default server install ? that having a default configuration suitable for workstation becomes a server problem ?
15:36:31 <sgallagh> #info Due to lack of quorum, FESCo will wait until next week to gather more information and make a decision.
15:36:50 <nirik> mclasen: is disabling autosuspend for live sessions something likely? or I guess I can ask in the Workstation meeting...
15:37:31 <mclasen> doesn't seem like a big deal to me. if it does active harm, we'll disable it
15:37:42 <sgallagh> mclasen: My gut says "no". This specifically I'm concerned about because I know it's a large enough userbase that we will get bugs on it.
15:37:53 <zbyszek> mclasen: I'd say that goes for any package. Installing a package not break other stuff. And we should always be very carefull on significant changes in behaviour on updates.
15:38:26 <sgallagh> mclasen: I mean, we can talk about adding `Conflicts: gdm` to fedora-release-server as a potential approach as well, but I think our users might find it annoying.
15:38:27 <nirik> yeah, the case of the usb cdrom drive failing to resume, and the suspending after install makes me think it's a good idea.
15:39:19 <mclasen> sgallagh: but there's absolutely no need for that
15:39:34 <mclasen> who are you protecting from what with such a conflicts line ?
15:40:07 <sgallagh> mclasen: I was throwing that out there as an exaggeration, sorry. That wouldn't be something I'd wish for.
15:41:05 <sgallagh> mclasen: In my ideal world, we'd have a config file somewhere with a true/false (or time value or whatnot) for the auto-suspend feature that Server could just push "disable" to
15:41:06 <zbyszek> sgallagh: I would be OK with your first proposal, i.e. kicking this back to Wksnt WG right now.
15:41:26 <zbyszek> sgallagh: I don't want to block this for another week, since the result is likely to be the same.
15:42:27 <sgallagh> zbyszek: Well, there are still only 4 voting members weighing in, so  a formal decision isn't possible today
15:42:31 <mclasen> sgallagh: I believe halfline explained how to do that. we can assist with working through the mechanics
15:42:42 <sgallagh> mclasen: Sure, let's do that.
15:42:57 <sgallagh> I want to be clear, I'm not trying to be hostile to the feature for Workstation.
15:43:04 <sgallagh> mclasen: Can we book an hour on Monday perhaps?
15:43:39 <mclasen> fine with me
15:43:47 <zbyszek> sgallagh: Yeah, but we could have #info, to keep things going, and then formally vote next week.
15:43:48 <sgallagh> OK, I'll send an invite to you and halfline
15:44:04 <zbyszek> But OK, it that's not needed, I'm happy to move on.
15:44:30 <sgallagh> #info FESCo will currently leave this decision for the Workstation WG, as long as a method not to impact Server is determined.
15:44:43 <sgallagh> #info sgallagh and mclasen will meet next week to discuss mitigations for Server
15:44:58 <zbyszek> ack
15:45:12 <sgallagh> #topic #1874 Package Anki has a Non-responsive Maintainer
15:45:12 <sgallagh> .fesco 1874
15:45:14 <zodbot> sgallagh: Issue #1874: Package Anki has a Non-responsive Maintainer - fesco - Pagure - https://pagure.io/fesco/issue/1874
15:45:20 <sgallagh> This seems pretty straightforward.
15:45:40 <maxamillion> I'm sorry, I have a conflict and must duck out
15:45:56 <sgallagh> maxamillion: No problem. This is the last ticket
15:46:00 <zbyszek> maxamillion: can you vote on this one, we don't have quorum
15:46:10 <sgallagh> There's +3 in the ticket already
15:46:12 * jsmith is finally here
15:46:17 <jsmith> Sorry I'm late
15:46:18 <sgallagh> I'm +1 to make it +4
15:46:27 <sgallagh> So nirik or maxamillion can finalize it
15:46:29 <jsmith> I was +1 in the ticket, and am still +1
15:47:02 <maxamillion> +1
15:47:08 <nirik> +1
15:47:50 <sgallagh> #agreed Anki package will be reassigned to thomasfedb (+6, 0, -0)
15:47:59 <sgallagh> nirik: Would you mind pulling the appropriate levers?
15:48:25 <nirik> sure, can do
15:49:02 <sgallagh> Do we have a list of Christian Krause's other packages and should we orphan them as well?
15:49:16 * sgallagh isn't sure how to look that up in the Pagure world
15:49:18 <nirik> good question
15:49:27 <nirik> https://src.fedoraproject.org/user/chkr
15:49:30 <nirik> 48 packages
15:49:45 <nirik> but they may be just co-maintainer on some
15:50:26 <sgallagh> Yeah, looks like a mix
15:50:34 <sgallagh> How do we want to handle this?
15:51:01 <sgallagh> Maybe orphan any for which he's the main admin or the sole contributor?
15:51:28 <sgallagh> (Sorry, that was two separate options)
15:51:59 <nirik> yeah, not sure.
15:52:07 <jsmith> I'd be fine with that option
15:53:23 <nirik> I can probibly figure out how to do that.
15:53:50 <nirik> but we should probibly see if we can get +5 in ticket before doing that
15:53:57 <nirik> since people didn't vote for that specifically.
15:53:57 <sgallagh> OK
15:54:29 <sgallagh> #info FESCo members are asked to vote in-ticket for what to do with the non-responsive maintainer's other packages
15:54:30 <jsmith> Sounds reasonable
15:54:41 <zbyszek> chrk is still active, at least tangentially
15:54:43 <zbyszek> https://bugzilla.redhat.com/show_bug.cgi?id=1563174#c1
15:54:53 <sgallagh> OK, I think that's everything on the agenda. In under an hour!
15:54:59 <sgallagh> #topic Next Week's Chair
15:55:09 * sgallagh pulls the pin
15:55:17 <sgallagh> Who is going to dive on it?
15:55:28 <zbyszek> So I think we shouldn't orphan all packages without a response and some more time.
15:55:30 <nirik> I will not be here next week.
15:55:34 <zbyszek> I can do it.
15:56:00 <sgallagh> #info zbyszek to run the 2018-04-13 FESCo meeting
15:56:11 <sgallagh> Happy Friday the 13th, zbyszek!
15:56:14 <sgallagh> (mwahahahaha)
15:56:19 <zbyszek> Do we need to discuss #1861?
15:56:31 <zbyszek> And what about #1871?
15:56:39 <zbyszek> #1877 sorry
15:57:11 <sgallagh> Oh, I missed 1861, didn't I
15:57:16 <zbyszek> sgallagh: thanks for the good wishes ;)
15:57:33 <sgallagh> #topic #1861 F28 Changes not in ON_QA status (<100% completed)
15:57:33 <sgallagh> .fesco 1861
15:57:35 <zodbot> sgallagh: Issue #1861: F28 Changes not in ON_QA status (<100% completed) - fesco - Pagure - https://pagure.io/fesco/issue/1861
15:58:15 <sgallagh> So, with zbyszek's last comment, I think this is in good shape.
15:58:20 <sgallagh> Anyone disagree?
15:58:58 <jsmith> Nope
15:59:09 <sgallagh> Let's call it done then
15:59:13 <nirik> +1
15:59:23 <zbyszek> +1
15:59:27 <sgallagh> #info FESCo is satisfied with the current state of the Changes
15:59:57 <sgallagh> zbyszek: I'd like to defer 1877 to next week, as it was filed late, we're already below quorum and we're about to lose nirik
16:00:07 <zbyszek> OK.
16:00:46 <sgallagh> #topic #1877 large number of packages FTBFS in F28
16:00:47 <sgallagh> .fesco 1877
16:00:50 <zodbot> sgallagh: Issue #1877: large number of packages FTBFS in F28 - fesco - Pagure - https://pagure.io/fesco/issue/1877
16:00:51 <jsmith> Let's try to drum up some discussion in the ticket in the meantime.
16:01:03 <sgallagh> #info Deferred to next week due to lack of quorum. Please discuss in the ticket.
16:01:08 <jsmith> No need to wait a week to start the discussion
16:01:10 <zbyszek> Also, since we don't have quorum, let's discuss the proposal to move the meeting time in the mail thread.
16:01:18 <sgallagh> zbyszek: Yes
16:01:23 <sgallagh> #topic Open Floor
16:01:25 <jsmith> WORKSFORME
16:01:38 * jsmith has nothing for the open floor
16:01:45 <nirik> I was going to mention one thing...
16:02:00 <sgallagh> #info FESCo members should discuss adjusting the meeting time for DST in an email thread
16:02:04 <sgallagh> nirik: Shoot
16:02:41 <nirik> It was pointed out recently that any maintainer can submit buildroot overrides (even if they dont have commits on the package(s)). IMHO this is fine, we corrected the docs... but if for some reason fesco wants a different behavior, we can adjust it.
16:03:31 <zbyszek> I think that's OK. It wasn't a problem until now apparently.
16:03:58 <nirik> just wanted to toss it out there.
16:04:09 <nirik> thats all I had (and I have to leave now anyhow)
16:04:12 <sgallagh> nirik: Are there protections on buildroot overrides of lower NVR?
16:04:36 <sgallagh> e.g. Can a non-maintainer intentionally stuff a GCC with a known vulnerability in code generation into the buildroot?
16:04:58 <nirik> not sure. It would definitely log who did it tho.
16:05:28 <sgallagh> nirik: I think it might be sufficient to make sure that pkg-owner@ just gets an email whenever an override of that package is created.
16:05:53 <jsmith> sgallagh: That sounds like a reasonable compromise
16:05:55 <nirik> if you could file that as a RFE I think that might be nice... or rather that it emits a fedmsg...
16:06:07 <sgallagh> ack. I'll do that
16:06:07 <nirik> (which it might already, dunno)
16:06:40 <sgallagh> #action sgallagh to file an RFE for buildroot overrides to always notify package owner(s) when they are created.
16:06:56 * jsmith has to run -- thanks everyone!
16:07:08 <sgallagh> Anyone have anything else today?
16:08:20 <sgallagh> Thanks for coming, folks
16:08:23 <sgallagh> #endmeeting