15:00:01 <contyk> #startmeeting FESCO (2019-03-11)
15:00:01 <zodbot> Meeting started Mon Mar 11 15:00:01 2019 UTC.
15:00:01 <zodbot> This meeting is logged and archived in a public location.
15:00:01 <zodbot> The chair is contyk. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:01 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:00:01 <zodbot> The meeting name has been set to 'fesco_(2019-03-11)'
15:00:04 <contyk> #meetingname fesco
15:00:04 <zodbot> The meeting name has been set to 'fesco'
15:00:06 <contyk> #chair nirik, bowlofeggs, jforbes, zbyszek, bookwar, sgallagh, contyk, mhroncok, otaylor
15:00:06 <zodbot> Current chairs: bookwar bowlofeggs contyk jforbes mhroncok nirik otaylor sgallagh zbyszek
15:00:08 <contyk> #topic init process
15:00:11 <jforbes> .hello2
15:00:12 <zodbot> jforbes: jforbes 'Justin M. Forbes' <jforbes@redhat.com>
15:00:13 <contyk> hello, party people
15:00:16 <contyk> .hello psabata
15:00:17 <zodbot> contyk: psabata 'Petr Šabata' <psabata@redhat.com>
15:00:20 <zbyszek> .hello2
15:00:21 <zodbot> zbyszek: zbyszek 'Zbigniew Jędrzejewski-Szmek' <zbyszek@in.waw.pl>
15:00:21 <bowlofeggs> .hello2
15:00:24 <zodbot> bowlofeggs: bowlofeggs 'Randy Barlow' <rbarlow@redhat.com>
15:00:47 <nirik> morning
15:00:48 <bowlofeggs> ain't no party like a fes-co party
15:00:49 <mhroncok> hello
15:00:58 <bowlofeggs> hay hoo
15:00:58 <otaylor> .hello2
15:00:59 <zodbot> otaylor: otaylor 'Owen Taylor' <otaylor@redhat.com>
15:01:00 <bookwar> .hello2
15:01:02 <zodbot> bookwar: bookwar 'Aleksandra Fedorova' <alpha@bookwar.info>
15:01:07 <nirik> are all us US folks saving some daylight? :)
15:01:28 <bowlofeggs> nirik: i've been stashing all my daylight away in an index fund
15:01:39 <zbyszek> bookwar: we forgot to ask if current meeting time is acceptable.
15:01:41 <contyk> so many people today!
15:01:58 <bookwar> zbyszek: works for me
15:01:59 <zbyszek> bookwar: if possible, I think we should continue using this slot
15:02:05 <zbyszek> OK, great.
15:02:17 <contyk> okay
15:02:32 <contyk> we don't have that many things today, which is great since this overlaps with the blocker meeting...
15:02:42 <contyk> let's start with the new business
15:02:44 <bcotton> .hello2
15:02:45 <zodbot> bcotton: bcotton 'Ben Cotton' <bcotton@redhat.com>
15:02:48 <bowlofeggs> you might even say this blocks the blocker meeting
15:02:51 <contyk> #topic #2101 Fedora 30 incomplete changes
15:02:53 <sgallagh> .hello2
15:02:54 <contyk> .fesco 2101
15:02:54 <zodbot> sgallagh: sgallagh 'Stephen Gallagher' <sgallagh@redhat.com>
15:02:56 <contyk> https://pagure.io/fesco/issue/2101
15:02:57 <zodbot> contyk: Issue #2101: Fedora 30 incomplete changes - fesco - Pagure.io - https://pagure.io/fesco/issue/2101
15:03:22 <mhroncok> contyk: one by one?
15:03:24 <nirik> shall we do them one at a time? or ?
15:03:29 <bowlofeggs> welp, these things aren't done, so i guess we should just cancel fedora 30
15:03:35 <bowlofeggs> yeah one at a time is good
15:03:43 <bcotton> bowlofeggs++
15:03:43 <zodbot> bcotton: Karma for bowlofeggs changed to 16 (for the current release cycle):  https://badges.fedoraproject.org/tags/cookie/any
15:03:50 <sgallagh> stransky_: Ping?
15:03:58 <stransky_> sgallagh, yes
15:04:01 <bowlofeggs> haha
15:04:04 <contyk> yeah, let's go one by one
15:04:23 <sgallagh> stransky_: https://fedoraproject.org/wiki/Changes/Firefox_Wayland_By_Default_On_Gnome has no update. Is it deferred?
15:04:24 <bowlofeggs> stransky_: how's https://fedoraproject.org/wiki/Changes/Firefox_Wayland_By_Default_On_Gnome going?
15:04:36 <mhroncok> it's written int he ticket
15:04:36 <bowlofeggs> jinx!
15:04:40 <zbyszek> sgallagh: it's waiting for F66
15:04:50 <stransky_> it's going well so far, week after ff66 should be ready for testing
15:05:09 <sgallagh> Oh sorry. I missed the last update there.
15:05:19 <mhroncok> my thinking: we have updates-testing enabled in beta so if we push this to updates-testing, we should get exposutre
15:05:24 <sgallagh> zbyszek: I assume you mean FF66 and not F66 :-D
15:05:30 <mhroncok> :D
15:05:31 <bowlofeggs> haha
15:05:32 * bcotton notes that FF66 is scheduled for a week before the beta target
15:05:46 <bowlofeggs> yeah this seems like a risky timeline
15:05:47 <sgallagh> bcotton: Yes, that's relevant.
15:05:51 <zbyszek> I'm testing F67.0a1, and it's come a long way, but there still are bugs...
15:05:58 <zbyszek> sgallagh: :)
15:06:12 <nirik> stransky_: possible to get a prerelease of it in sooner? or not worth it?
15:06:26 <bcotton> i think the question is: if we were to land it immediately after the beta freeze, is that too late?
15:06:39 <stransky_> yes, I'm working on the last importnat ones now. and yes, I can backport to 65
15:06:44 <sgallagh> bcotton: Aren't we *in* Beta Freeze?
15:06:52 <nirik> that does reduce testing somewhat...
15:06:55 <bcotton> sgallagh: yes, i mean after it thaws
15:07:02 <mhroncok> this is going to be in updates-testing anyway, enabled by default, right?
15:07:11 <stransky_> if it's impornant it can be decided sooner, right after FF66 update
15:07:11 <nirik> stransky_: when do you think you might have a build ready?
15:07:15 <mhroncok> what's the difference in the testing amount it gets?
15:07:29 <otaylor> bcotton: sooner the better, IMO. I don't think there's much risk of this delaying the beta
15:07:32 <sgallagh> mhroncok: Many people only test the Live
15:07:32 <nirik> mhroncok: some people boot beta live media and test things and then move on... they never apply updates
15:07:47 <jforbes> some
15:07:48 <stransky_> if you say it has to be in FF66 release time the builds will be ready
15:07:55 <stransky_> for testing
15:07:58 <mhroncok> in that case wee need an exception to make this land in beta, right?
15:08:06 <sgallagh> mhroncok: Yes
15:08:07 <nirik> yes, we already do.
15:08:17 <sgallagh> Or declare it a "FESCo Blocker"
15:08:20 <mhroncok> just making sure
15:08:21 <sgallagh> But I'm not prepared to do that
15:08:48 <bcotton> speaking for myself, i'm not particularly keen on landing a change in a default, widely-used package this late in the process. but also we do it a lot so shrug
15:09:10 <contyk> it's just a web browser
15:09:17 <mhroncok> exactly :D
15:09:25 <sgallagh> stransky_: The fallback plan is going back to the X version for Final, right?
15:09:32 <nirik> I'd be ok with leaving it to the FE process... the sooner there's a build to propose the more likely it would be to get approved...
15:09:32 <stransky_> sgallagh, sure
15:09:59 <mhroncok> proposal...
15:10:07 <stransky_> I'll make sure the beta gets usable browser
15:10:11 <stransky_> of course
15:10:17 <sgallagh> Proposal: if the F66 is ready at least before the first RC of Beta, it can be default
15:10:33 <zbyszek> sgallagh: FF
15:10:35 <nirik> well, or a build with the needed stuff backported?
15:10:35 <mhroncok> stransky_: how exactly
15:10:39 <contyk> what do you mean by ready? ;)
15:10:39 <sgallagh> heh, yes
15:10:50 <mhroncok> stransky_: we either put it in beta for testsers
15:10:55 <sgallagh> Built and in Bodhi
15:10:55 <mhroncok> stransky_: or not
15:10:57 <bowlofeggs> sgallagh: s/F66/the change/ ?
15:11:07 <stransky_> there are known bugs which need to be fixed
15:11:10 <sgallagh> bowlofeggs: Sure
15:11:24 <bowlofeggs> sgallagh: +1 from me
15:11:31 <zbyszek> So there are two issues: whether to update to FF66 (which I think is pretty safe), and whether to switch go wayland by default (which is much more risky...)
15:11:33 <nirik> well, so that bypasses the normal process?
15:11:41 <sgallagh> Patch Proposal: "If the Firefox Wayland Change is ready for testing in time for the first RC of Beta, it can go in"
15:11:56 <zbyszek> sgallagh: +1
15:11:58 <contyk> +1 to the patched proposal
15:12:02 <bowlofeggs> sgallagh: +1
15:12:03 <jforbes> +1
15:12:05 <mhroncok> sgallagh: +1
15:12:06 <bookwar> +1
15:12:30 <nirik> who decides it's ready?
15:12:40 <zbyszek> I think the right contingency plan will be to simply switch the default, but keep FF66.
15:12:42 <sgallagh> nirik: Change proposer
15:12:48 <stransky_> BZ tracked I guess is the first instance :)
15:12:49 <sgallagh> zbyszek: Yes
15:12:51 <bowlofeggs> zbyszek: +1
15:12:52 <stransky_> tracker
15:12:55 <otaylor> +1 (though I don't actually know the RC schedule)
15:12:56 <contyk> zbyszek: yeah
15:12:57 <nirik> counter proopsal: this change uses the normal FE process and if approved before beta can go in
15:13:11 <sgallagh> otaylor: The RC schedule is "when the known blockers are cleared"
15:13:16 <nirik> then QA and others can look and decide
15:13:32 <mhroncok> nirik: I assumed it needs an excetpion to be ready to beta
15:13:46 <mhroncok> if there is no exception / blocker, it cannot go to beta and hence is not ready
15:13:49 <sgallagh> nirik: The FE process can decide that "they would consider".
15:14:03 <bowlofeggs> nirik: yeah i guess i assumed it was implied in sgallagh's proposal too - we could make it explicit though - +1
15:14:23 <sgallagh> My proposal is "FESCo wants this in, if it's done in time"
15:14:47 <nirik> perhaps I am nitpicking, but the proposal sounds like 'if the change owner says it's ok, it bypasses FE and blockers and goes directly in'
15:14:50 <mhroncok> sgallagh: it wasn't done in time
15:15:10 <nirik> which I don't think is a good way to do things.
15:15:16 <mhroncok> certainly not
15:15:30 <bowlofeggs> yeah i want it to go through the FE/blocker process
15:15:38 <sgallagh> OK, let me try one more patch
15:15:55 <bookwar> nirik: i think we assumed the same thing as you've written, but your wording is better indeed
15:16:23 * nirik waits for sgallagh
15:16:26 <sgallagh> Proposal: FESCo would like to see this Change land if it would not delay the Beta release. The standard blocker/exception review will make the final call.
15:16:49 <nirik> sure, +1 and thanks.
15:16:56 <contyk> ack
15:16:59 <jforbes> +1
15:16:59 <mhroncok> ...if it doesn't land in beta, it is deferred to F31
15:17:01 <mhroncok> right?
15:17:04 <sgallagh> mhroncok: Yes
15:17:07 <mhroncok> +1
15:17:13 <otaylor> +1
15:17:15 <zbyszek> sgallagh, nirik: does this mean that FESCo asks for this update to be included in F30 beta?
15:17:19 <contyk> #info Firefox Wayland by default on GNOME -- FESCo would like to see this Change land if it would not delay the Beta release. The standard blocker/exception review will make the final call.
15:17:23 <mhroncok> zbyszek: no
15:17:24 <bowlofeggs> sgallagh: +1
15:17:31 <sgallagh> Proposal: FESCo would like to see this Change land if it would not delay the Beta release. The standard blocker/exception review will make the final call. If it does not land in F30 Beta, this Change is Deferred to F31.
15:17:41 <sgallagh> Just to be clear what we're agreeing to for the record.
15:17:47 <contyk> okay
15:17:48 <nirik> zbyszek: asks to be considered I guess?
15:17:50 <contyk> #undo
15:17:50 <zodbot> Removing item from minutes: INFO by contyk at 15:17:19 : Firefox Wayland by default on GNOME -- FESCo would like to see this Change land if it would not delay the Beta release. The standard blocker/exception review will make the final call.
15:17:52 <bowlofeggs> sgallagh: +1
15:18:02 <contyk> #info Firefox Wayland by default on GNOME -- FESCo would like to see this Change land if it would not delay the Beta release. The standard blocker/exception review will make the final call. If it does not land in F30 Beta, this Change is Deferred to F31.
15:18:08 <contyk> next one
15:18:24 <zbyszek> contyk: wait a moment please
15:18:34 * nirik has to step away for a sec
15:19:16 <zbyszek> I'm confused, because we're in freeze now. So this change will NOT go in, because there's no RC bug to fix.
15:20:14 <zbyszek> "During these times, builds will not be marked stable and moved from updates-testing to fedora (and hence included in the milestone release composes) except for those approved under the QA:SOP_blocker_bug_process or QA:SOP_freeze_exception_bug_process."
15:20:47 <contyk> isn't the second sentence assuming the FE process?
15:20:50 <jforbes> It was said that an exception would need to be filed
15:20:56 <sgallagh> zbyszek: I have just proposed it for FE
15:21:42 <zbyszek> https://fedoraproject.org/wiki/QA:SOP_freeze_exception_bug_process#Freeze_exception_bug_principles
15:22:16 <zbyszek> It seems to be me that a big default browser update is nowhere near any of the listed cases.
15:22:19 <mhroncok> the fix is reasonably small and testable
15:22:30 <mhroncok> as contyk said, it's just a webbrowser
15:22:32 <sgallagh> zbyszek: The review starts in 40 minutes in #fedora-blocker-review
15:22:46 <contyk> ;)
15:22:48 <sgallagh> If you want to argue that it should not be granted an exception, do so there, please
15:22:54 <zbyszek> OK, I'm not convinced, but let's move on.
15:23:00 <contyk> thanks
15:23:03 <mhroncok> one question
15:23:22 <mhroncok> do we say: we want this to be granted an exception? or do we say: we want the usual process to happen?
15:23:24 <zbyszek> sgallagh: I think it *should* be granted an exception, but as a FESCo priority.
15:23:39 <contyk> mhroncok: I think we want the usual process to happen
15:23:49 <contyk> if it's not granted an FE, the change will not land
15:23:49 <mhroncok> ok
15:24:04 <contyk> ok, Make BootLoaderSpec-style configuration files the default?
15:24:07 <sgallagh> The usual process should happen, but FESCo is indicating that we consider this valuable to the distro (but not enough that we want to block on it)
15:24:31 <zbyszek> sgallagh: OK, but can we make this a formal statement?
15:24:36 <sgallagh> contyk: There was some chatter about this in the Workstation meeting this morning
15:24:47 <sgallagh> zbyszek: I think we did
15:25:10 <sgallagh> "FESCo would like to see this Change land if it would not delay the Beta release."
15:25:15 <sgallagh> Can we please move on?
15:25:23 <zbyszek> Yes, please do.
15:25:37 <bowlofeggs> the bootloader one got set to ON_QA
15:25:38 <otaylor> On the next incomplete change - The BLS change was complete (according to the owners), just not moved to ON_QA - I got javierm to move it this morning
15:25:39 <jforbes> sgallagh: what was the chatter this morning?
15:25:42 <bowlofeggs> so i think we can move on
15:25:57 <jforbes> So it sounds like it is on track?
15:26:01 <contyk> #info Make BootLoaderSpec-style configuration files the default should be ready now
15:26:07 <contyk> Haskell GHC 8.4 and Stackage LTS 12
15:26:14 <sgallagh> jforbes: I heard mention of upgrades having issues, but no specific examples
15:26:18 <mhroncok> that one makes me sad
15:26:26 <sgallagh> mhroncok: Which?
15:26:31 <mhroncok> Haskell GHC 8.4 and Stackage LTS 12
15:26:36 <nirik> so... IMHO we cannot just tag this in now.
15:26:43 <nirik> it has to be an update and go via the process.
15:26:49 <nirik> but it's not on any media I don't think
15:27:12 <contyk> what's the issue?
15:27:12 <sgallagh> They set their contingency deadline at *branch*, not Freeze, either.
15:27:21 <bowlofeggs> sounds like we are waiting on the packager in that releng ticket?
15:27:33 <nirik> well, now, but I just commented...
15:28:01 <mhroncok> fairly the thing was requested before the beta freeze
15:28:07 <mhroncok> quite late
15:28:22 <nirik> yeah.
15:28:27 <mhroncok> Contingency deadline: Before branching of F30
15:28:41 <mhroncok> this should have been reverted
15:28:43 <nirik> juhp: you around?
15:28:54 <sgallagh> Is reverting an option at this point?
15:29:00 <contyk> it's in a side tag
15:29:03 <contyk> so yes
15:29:12 <sgallagh> Then I think we should defer it.
15:29:13 <otaylor> nirik: 12:30am in japan
15:29:22 <nirik> yeah, true.
15:29:27 <sgallagh> Proposal: defer haskell update to F31
15:29:32 <contyk> +1
15:29:37 <jforbes> +1
15:29:38 <bowlofeggs> sgallagh: +1
15:29:46 <mhroncok> 0
15:30:11 <nirik> +1 I guess... but it's such a leaf I don't think there would be much harm letting it land in f30...
15:30:12 * mhroncok has another proposal
15:30:21 <contyk> #info Haskell GHC 8.4 and Stackage LTS 12 -- past contingency deadline, deferring to F31
15:30:28 <zbyszek> -1, for the reasons that nirik stated
15:30:30 * mhroncok waits for this vote to finish
15:30:37 <bookwar> defer or leave for component owner to deal with updates?
15:30:58 <contyk> I'd rather not push such a big change as updates at this point
15:31:17 <sgallagh> same
15:31:29 <mhroncok> proposal: put everything to a large bodhi update, let it go in after beta. it's a giant leaf
15:31:40 <zbyszek> mhroncok: +1
15:31:47 <sgallagh> Of course, we're probably going to rubber-stamp the usual post-Freeze GNOME updates, so I feel kind of hypocritical :-/
15:31:49 <mhroncok> we do larger updates in the middle of stable release
15:32:01 <nirik> mhroncok: +1
15:32:03 <mhroncok> exactly
15:32:08 <contyk> 0
15:32:43 <jforbes> I could see an argument for it being pushed as an update honestly
15:33:19 <contyk> alright
15:33:22 <contyk> #undo
15:33:22 <zodbot> Removing item from minutes: INFO by contyk at 15:30:21 : Haskell GHC 8.4 and Stackage LTS 12 -- past contingency deadline, deferring to F31
15:33:23 <sgallagh> 0
15:33:42 <bowlofeggs> i'm ok with it being an update
15:33:44 <sgallagh> (I don't like it, but not enough to stand in the way)
15:33:50 <bookwar> i feel like managing it as a "giant leaf" essentially equivalent to letting update go through after release
15:34:03 <bowlofeggs> i would prefer a deferral, but i would approve it as an update too
15:34:05 <contyk> #info Haskell GHC 8.4 and Stackage LTS 12 -- if the maintainer wants this change to land in F30, they need to push it as an update bundle, otherwise defer to F31
15:34:19 <contyk> next one?
15:34:23 <contyk> MongoDB removal
15:34:23 <zbyszek> contyk: ack
15:34:38 <mhroncok> mongodb removal is funny ony
15:34:51 <mhroncok> if we say we deffer it, are we goign to add it back? probably not
15:35:49 <zbyszek> I think the remaining deps should be treated as normal bugs.
15:35:50 <mhroncok> yet the change owner is not very communicative, I remember the proposal being very brief and franky they just removed donẗ care about dependent packages
15:35:59 <nirik> so, those packages... is the solution to remove them? or ?
15:36:13 <sgallagh> mhroncok: Well, at the same time, upstream is actively hostile to distros :-/
15:36:19 <mhroncok> I agree
15:36:20 <nirik> zbyszek: yeah, that seems reasonable.
15:36:28 <bookwar> there should have been a "Contingency mechanism": if packages were not adapted - they are removed as well
15:36:39 <sgallagh> I can understand not wanting to deal with extra bureaucracy to STOP maintaining it
15:36:39 <mhroncok> I guess I'd +1 what zbyszek said
15:36:42 <bowlofeggs> i think we can just let FTBFS take it from here
15:37:11 <contyk> I think so too
15:37:11 <sgallagh> zbyszek: Make that a proposal?
15:37:15 <mhroncok> bowlofeggs: runtime deps
15:37:28 <mhroncok> fawkes-devenv
15:37:29 <otaylor> bowlofeggs: they may not be FTBFS, they may require mongodb as a runtime thing.. probably against fedora policy to runtime dep something not in the distro, but no mechanism there
15:37:41 <zbyszek> mongo-tools is FTBFS
15:37:42 <mhroncok> otaylor: I'm wornking on it
15:37:47 <mhroncok> *working
15:38:04 <mhroncok> fawkes is FTBFS
15:38:14 <mhroncok> ok, zbyszek, make it a proposal
15:38:28 <zbyszek> proposal: the Change is implemented, but some dependencies on removed packages remain. Those should be treated as bugs and resolved in the usual fashion.
15:38:32 <bowlofeggs> yeah if they require it as a runtime we have no choice but removal
15:38:38 <bowlofeggs> but these did show mongo as a build dep
15:38:39 <mhroncok> zbyszek: +1
15:38:46 <bowlofeggs> zbyszek: +1
15:38:49 <contyk> +1
15:38:58 <otaylor> +1
15:39:01 <mhroncok> bowlofeggs: fawkes-devenv-0:1.0.1-18.fc29.x86_64 is runtime, mongo-tools-0:4.0.4-4.20181124git0f0d866.fc30.src is build time
15:39:03 <sgallagh> +1
15:39:06 <jforbes> _1
15:39:08 <jforbes> +1 even
15:39:10 <bookwar> +1
15:39:21 <bowlofeggs> mhroncok: right, but isn't fawkes runtime *and* buildtime, meaning that FTBFS will find it anyway?
15:39:24 <nirik> +1
15:39:30 <sgallagh> jforbes: _1 is the most ambiguous typo you could have made :-P
15:39:37 <jforbes> indeed
15:39:39 <contyk> #info MongoDB removal -- the Change is implemented, but some dependencies on removed packages remain. Those should be treated as bugs and resolved in the usual fashion.
15:39:46 <contyk> glibc32 build adjustments
15:40:07 <mhroncok> Categories: ChangeIncomplete
15:40:20 <nirik> I think this is done, but looking
15:40:32 <otaylor> contyk: Looks like the releng change was done, but don't know if there were followup packaging changes after that
15:40:34 <contyk> there isn't much in the bug
15:40:37 <mhroncok> bcotton: is that category a mistake?
15:40:47 <otaylor> https://pagure.io/pungi-fedora/pull-request/636 was the releng change that blocked it going in for F29
15:41:14 <bcotton> mhroncok: category for what?
15:41:28 <mhroncok> bcotton: https://fedoraproject.org/wiki/Changes/glibc32_Build_Adjustments
15:41:42 <nirik> yeah, the releng part is done...
15:41:52 <bcotton> mhroncok: yes it is
15:41:54 <bcotton> fixing
15:41:57 <mhroncok> good
15:42:33 <sgallagh> Given that this is a build-time only issue, I'm probably fine with landing this whenever it's done.
15:42:35 <mhroncok> https://src.fedoraproject.org/rpms/glibc32/commits/master
15:43:00 <nirik> yeah, it didn't get retired...
15:43:15 <nirik> so it's not all done
15:43:25 <mhroncok> I wouldn't change it at this point on f30 really
15:43:33 <mhroncok> proposal: move to f31
15:43:41 <nirik> +1
15:43:46 <bowlofeggs> mhroncok: +1
15:43:53 <contyk> +1
15:44:14 <otaylor> +1
15:44:26 <bookwar> +1
15:44:27 <zbyszek> +1
15:44:37 <jforbes> +1
15:44:42 <contyk> #info glibc32 build adjustments -- not ready, deferring to F31
15:44:51 <contyk> Build non-RELRO ELF binaries with .got.plt isolation
15:45:22 <mhroncok> contyk: 1 moment
15:45:27 <nirik> same here I think
15:45:55 <mhroncok> additional proposal for glibc32: there was no reply since we moved this to f30. needinfo the change owner to ack it as f31 change. if there is no reply until we evaluate this for f31, we cancel the change
15:46:12 <bcotton> mhroncok: +1
15:46:16 <nirik> sure. +1
15:46:23 <zbyszek> +1
15:46:23 <jforbes> +1
15:46:26 <contyk> ack
15:46:29 <contyk> #undo
15:46:29 <zodbot> Removing item from minutes: INFO by contyk at 15:44:42 : glibc32 build adjustments -- not ready, deferring to F31
15:46:29 <otaylor> +1
15:46:35 <bowlofeggs> mhroncok: +1
15:46:44 <contyk> #info glibc32 build adjustments -- not ready; there was no reply since we moved this to f30. needinfo the change owner to ack it as f31 change. if there is no reply until we evaluate this for f31, we cancel the change
15:47:03 <contyk> I think the next one will be the same
15:47:29 <mhroncok> +1 for the same proposal
15:47:38 <zbyszek> +1 for the same resolution
15:47:39 <bowlofeggs> yeah +1 for same
15:48:13 <bookwar> +1
15:48:17 <otaylor> +1
15:48:19 * bcotton has updated the glibc32 change page and BZ tracker
15:48:23 <sgallagh> +1
15:48:26 <jforbes> +1
15:48:39 <contyk> #info Build non-RELRO ELF binaries with .got.plt isolation -- also not ready; there was no reply since we moved this to f30. needinfo the change owner to ack it as f31 change. if there is no reply until we evaluate this for f31, we cancel the change
15:48:44 <contyk> Reset locale if not available
15:48:53 * zbyszek is here
15:49:07 <mhroncok> zbyszek: I really want this
15:49:25 <zbyszek> So... it's a tiny change, and I hope it can still go in for F30.
15:49:27 <mhroncok> zbyszek: is there a technical problem or just not enough time to finish in time?
15:49:29 <jforbes> I see that it is delayed, but why is it delayed?
15:49:44 <zbyszek> ovasik said he'll build for F30.
15:50:00 <zbyszek> I made the implemnetation late, my fault.
15:50:10 <contyk> would that be an update or FE?
15:50:11 * bcotton has updated the non-relro change page and BZ tracker
15:50:44 <zbyszek> I'll ask for a FE.
15:50:54 <contyk> sounds acceptable to me
15:50:55 <mhroncok> since this affects containers and such, I think we need to stick it in beta to actaully test it
15:51:00 <sgallagh> Anything that changes the build flags should be scheduled ahead of a mass-rebuild, shouldn't they?
15:51:09 <sgallagh> Especially ones for security benefits.
15:51:13 <mhroncok> sgallagh: build flags?
15:51:38 * contyk doesn't think this is about build flags
15:51:47 <nirik> sgallagh: off by one error... we are talking about reset locale now. ;)
15:51:50 <sgallagh> maybe I'm misunderstanding
15:51:55 <sgallagh> gah
15:52:08 * sgallagh in two meetings at once. Must have missed one
15:52:10 <mhroncok> ok, same resolution as firefox?
15:52:26 <nirik> mhroncok: seems reasonable
15:52:28 <contyk> +1
15:52:28 <zbyszek> mhroncok: ack, +1
15:52:30 <mhroncok> zbyszek: can you please push this to rawhide at least ASAP? do you need help?
15:52:59 <zbyszek> mhroncok: I didn't want to use pp privs to push this. I'll ask the maintainer again.
15:53:00 <bookwar> agree that we need it to be in beta, locale bugs are often most simple and most embarrassing at the same time :)
15:53:35 <contyk> #info Reset locale if not available -- code ready but there's no build yet; we'll follow the standard fe/blocker process; if it doesn't land by beta, it will be deferred to f31
15:53:41 <mhroncok> zbyszek: if you prep a src.fp.o PR, I'll kick ovasik to merge it
15:53:45 <zbyszek> bcotton: should I open a new bug, or can I use the change tracking bug to nominate for FE?
15:53:47 <contyk> okay~
15:54:00 <zbyszek> mhroncok: will do
15:54:19 <contyk> we have six minutes; do we even want to start with the followups?
15:54:33 <nirik> zbyszek: probibly a new one to avoid closing or messing with the change bug?
15:54:45 <zbyszek> nirik: ok
15:54:50 <bowlofeggs> don't we have 1 hour and 6 minutes?
15:54:53 <nirik> IMHO at least.
15:54:54 <zbyszek> contyk: yeah, let's use the time.
15:55:09 <contyk> bowlofeggs: I think the blocker meeting is here
15:55:14 <bowlofeggs> ah
15:55:15 <contyk> okay
15:55:18 <contyk> #topic #2096 F31 System-Wide Change: BuildRequires generators
15:55:20 <contyk> .fesco 2096
15:55:21 <nirik> nope... they have their own channel.
15:55:21 <zodbot> contyk: Issue #2096: F31 System-Wide Change: BuildRequires generators - fesco - Pagure.io - https://pagure.io/fesco/issue/2096
15:55:22 <contyk> https://pagure.io/fesco/issue/2096
15:55:28 <contyk> so I added my comment to this one
15:55:41 <bcotton> zbyszek: you can probably use the tracking bug for that. adamw will tell us if you did it wrong :-)
15:55:46 <contyk> I think it'd be a nice feature but I'd rather improve the SRPM situation than make it worse
15:55:47 <mhroncok> contyk: I hear your comments and I agree. do you think we can summarize it as a hard requirement?
15:56:17 <contyk> well
15:56:54 <contyk> ignatenkobrain's is also valid but you can regenerate the SRPMs quite easily
15:57:08 <contyk> I'm not sure if that would be possible in this scenario; I suppose I could formulate it somehow...
15:57:15 <jforbes> I might suggest, there is still very active and useful discussion in ticket, and this is an f31 change, so not exactly timing critical. Perhaps defer a week instead of waiting for everyone to catch up on a book?
15:57:27 <nirik> there's a lot of activity on this this morning which I haven't had time to digest...
15:57:31 <ignatenkobrain> .hello2
15:57:32 <zodbot> ignatenkobrain: ignatenkobrain 'Igor Gnatenko' <i.gnatenko.brain@gmail.com>
15:57:36 <zbyszek> jforbes: agreed
15:57:36 <nirik> jforbes: +1
15:57:48 <contyk> ignatenkobrain: got to it in time ;)
15:58:14 <ignatenkobrain> I've got exit game posponed by 15 mins, so I have some time to answer any questions if you have
15:58:22 <contyk> ignatenkobrain: we need to end this in one
15:58:36 <contyk> so yeah, I'd also postpone this and continue in the ticket
15:58:53 <contyk> I can define my SRPM requirements more formally in the ticket and perhaps we can come up with some more things we'd want
15:59:25 <contyk> ok? ok
15:59:41 <contyk> #info We'll continue the discussion in the ticket for another week.
16:00:00 <ignatenkobrain> we don't have any production-ready implementation, we are still discussing implementation design
16:00:08 <contyk> good!
16:00:09 <ignatenkobrain> so it is right time to raise your concerns
16:00:25 <ignatenkobrain> and I'll be happy to make sure they are resolved before we get real impl
16:00:41 <contyk> ah, the blocker review isn't here
16:00:44 <contyk> so I think we can continue
16:01:13 <contyk> #topic #2092 Fedora 31 schedule
16:01:16 <contyk> .fesco 2092
16:01:17 <zodbot> contyk: Issue #2092: Fedora 31 schedule - fesco - Pagure.io - https://pagure.io/fesco/issue/2092
16:01:18 <contyk> https://pagure.io/fesco/issue/2092
16:01:25 <adamw> bcotton: what what what'd i do
16:02:14 <bcotton> adamw: the question was whether to submit an FE using the change tracker bug or if a new bug should be created
16:02:24 <mhroncok> the schedule is technically approved already, right?
16:02:59 <adamw> bcotton: i wouldn't use the change tracker bug, because to close out the FE process we have to actually close the bug, but the change tracker bug shouldn't necessarily be closed just for whatever work occurs at beta, right?
16:03:06 <adamw> we would still want the tracker bug open at final
16:03:13 <bcotton> ooh, good point. zbyszek^^
16:03:15 <zbyszek> adamw: thanks, I'll open a new bug.
16:03:19 <adamw> npnp
16:03:38 <zbyszek> mhroncok: counting you as +1, we're at (+4, 0, 0)
16:03:50 <jforbes> seems suboptimal, but okay
16:03:54 <bcotton> mhroncok: i'm acting like it is, but the schedule can always be changed later if need be :-)
16:04:43 <mhroncok> zbyszek: yes, I'm +1
16:05:19 <nirik> +1 to schedule
16:05:51 * zbyszek is still +1
16:05:57 <otaylor> +1 to schedule
16:06:05 * jforbes is still +1
16:06:34 * contyk is rather distracted now
16:06:44 <contyk> +1
16:07:04 <contyk> so that's... +6?
16:07:39 <bookwar> there are issues with python, is KDE release also a problem for that new release?
16:07:48 <kalev> I have a quick note on this topic
16:07:58 <mhroncok> bookwar: there are issues with python?
16:08:13 <bookwar> with python release date :)
16:08:14 <kalev> GNOME 3.34 schedule was just published and 3.34.1 release is two days after proposed F31 final freeze
16:08:26 <mhroncok> heh
16:08:33 <mhroncok> everyhting is a bit later this time
16:08:46 <kalev> would be slightly better to have F31 final freeze start 1 week later, but I suppose we can manage with a late FE too if it can't be done
16:08:51 <mhroncok> and we cannot pospone f31 because holidays :(
16:08:51 <bookwar> maybe it is us who are too early? )
16:09:02 <zbyszek> Let's move the holidays.
16:09:10 <zbyszek> I think Canada has them at a later date.
16:09:32 <mhroncok> zbyszek: sure, at least we would maybe get some snow on xmas day in Prague
16:10:06 <bookwar> move to orthodox calendar..
16:10:13 <bcotton> "cannot" is a strong statement, but postponing the fall release starts to get tricky very quickly
16:11:31 <bookwar> bcotton: currently we are not postponing explicitly but have big things land later, which means we are violating the schedule already
16:12:22 <kalev> bcotton: can you add GNOME 3.34.1 to the F31 schedule please? Wed 2019-10-09
16:12:27 <bcotton> kalev: ack
16:12:30 <kalev> thanks
16:14:05 <bcotton> bookwar: the difference is that landing things late may or may not push back dates. delaying the schedule means that slippages get us into the holidays and we run the risk of not having the people we need around to get the release out the door
16:14:46 <kalev> anyway, the GNOME schedule conflict isn't the end of the world and if it's difficult to move the schedule we can work with that, just a bit more work for developement and QA to test the late freeze breaks
16:14:46 <contyk> ok, back
16:15:12 <zbyszek> wow, that list long...
16:16:13 <zbyszek> Anyway, anything to do here?
16:16:32 <contyk> sounds like there are some new items we need to consider and come up with a new schedule?
16:16:35 <contyk> if I'm reading it right
16:17:02 <bcotton> the schedule is approved right now based on FESCo's voting policy
16:17:23 <bookwar> so Gnome3.14.1 and Python3.8 are treated as exceptions
16:17:29 <bookwar> or not?
16:17:33 <zbyszek> contyk: I'd rather say "there are some external events which don't align well with our schedule, so we'll need to shoehorn some packages into the schedule."
16:17:35 <mhroncok> there is no gnome change yet
16:17:37 <bcotton> so if someone wants a new schedule, they should tell me what they want and then we can decide if it's worth it
16:17:46 <contyk> okay, thought we were waiting for that python thing
16:18:13 <bcotton> contyk: mhroncok was, but other people voted without him
16:18:16 <nirik> if we can adjust in advance for those things I think that would be better than landing them after the freeze...
16:18:22 <contyk> okay, let me review the count
16:19:15 <contyk> okay
16:19:36 <contyk> #agreed Fedora 31 schedule is APPROVED (+7, 0, -0)
16:19:49 <contyk> #topic #2027  RFC: Module lifecycles
16:19:52 <contyk> .fesco 2027
16:19:54 <contyk> https://pagure.io/fesco/issue/2027
16:19:54 <zodbot> contyk: Issue #2027: RFC: Module lifecycles - fesco - Pagure.io - https://pagure.io/fesco/issue/2027
16:19:56 <mhroncok> if we move the release to January 2020, that would actually give some ammunition to the Python 2 deletionists :D
16:19:56 <contyk> sgallagh, asamalik ^
16:20:03 <asamalik> .hello2
16:20:04 <zodbot> asamalik: asamalik 'Adam Samalik' <asamalik@redhat.com>
16:20:07 <bcotton> mhroncok++
16:20:07 <zodbot> bcotton: Karma for churchyard changed to 10 (for the current release cycle):  https://badges.fedoraproject.org/tags/cookie/any
16:20:09 <mhroncok> the new propsoal landed jjust before the meeting
16:20:14 <mhroncok> I wasn't able to read it yet
16:20:15 * asamalik waves
16:20:37 <zbyszek> I'm not up-to-date either
16:20:40 <contyk> me neither
16:20:50 * jforbes either
16:21:02 <asamalik> This has been discussed (I think) two week ago and rejected by FESCo. We have discussed it on the Modularity WG meeting and sgallagh updated it.
16:21:21 <asamalik> Yeah it's been a last-minute update...
16:21:57 <sgallagh> Sorry about that. I forgot to do it last week.
16:21:58 <contyk> I'd propose giving FESCo another week to review it
16:21:58 <jforbes> So why are we discussing it right now?
16:22:06 <contyk> but if you can summarize the changes now, that would also help
16:22:19 <zbyszek> https://pagure.io/modularity/working-documents/c/dcd4d53df39b2500e19ad67b323fd90972046d94
16:22:42 <sgallagh> The short version is we stopped trying to dictate an implementation, since that’s where FESCo got hung up.
16:22:45 * contyk hates horizontal scrolling
16:23:08 <sgallagh> We now state that it must have a public API to retrieve the EOL data.
16:23:16 <sgallagh> But not what must provide that API
16:23:20 <asamalik> Basically leaving out unnecessary implementation details, and rather stating requirements
16:23:26 <asamalik> that :)
16:23:40 <contyk> ok to give it another week then?
16:23:48 <asamalik> contyk: sure
16:24:28 <contyk> #info FESCo will review the new proposal within the following week
16:24:45 <contyk> alright
16:24:49 <contyk> #topic Next week's chair
16:24:50 <mhroncok> "these values may be extended (enabling support on later releases), but may not be shortened."
16:24:55 <mhroncok> that's creepy
16:25:08 <mhroncok> anyway, will look trough it next week
16:25:14 <contyk> any volunteers?
16:25:36 * mhroncok might not show up at all, not sure yet
16:25:37 <jforbes> I can do it
16:25:41 <mhroncok> jforbes++
16:25:44 <contyk> jforbes++
16:25:46 <zodbot> contyk: Karma for jforbes changed to 7 (for the current release cycle):  https://badges.fedoraproject.org/tags/cookie/any
16:25:52 <contyk> #action jforbes will chair next meeting
16:25:57 <contyk> #topic Open Floor
16:26:35 <bookwar> do we have a full list of things which are risky for current schedule (like Python3.8 and Gnome)? do we plan to build this list via Accepted Changes worklow?
16:27:39 <mhroncok> I suppose that if somebody proposes a change for a new gnome  version, they should provide the data points, plan and contingency if it doesn't release before finalf reeze
16:28:57 <bookwar> for Gnome it is already after the freeze
16:29:18 <mhroncok> sure, so the change proposal must be accompanied with what's the actual plan
16:29:38 <bookwar> ok, so we would discuss it through the change acceptance
16:29:42 <nirik> there's a pending update and a FE bug...
16:30:28 <bookwar> feels a bit weird to approve release schedule before knowing what is going to happen, but seems there is no much freedom anyway
16:30:43 <bookwar> ok, i am done with my question, thanks
16:31:34 * contyk nods
16:31:41 <contyk> anything else for the open floor? closing in a minute if not
16:32:40 <contyk> alright
16:32:44 <contyk> #endmeeting