[15:00] <infinity> o/
[15:01] <sil2100> o/
[15:04] <cyphermox> o/
[15:04] <rbalint> o/
[15:04] <cyphermox> yay meeting!
[15:05] <tdaitx> hmm, steve is away?
[15:09]  * xnox wonders if i can reboot back into applying windows updates
[15:13] <cyphermox> let's have the meeting anyway?
[15:13] <rbalint> i would love to talk about reviewing the current Ubuntu delta for packages
[15:13] <infinity> cyphermox: Is that you volunterring to run it?
[15:13] <cyphermox> yeah
[15:13] <cyphermox> #startmeeting foundations
[15:13] <meetingology> Meeting started Thu Apr 27 15:13:25 2017 UTC.  The chair is cyphermox. Information about MeetBot at http://wiki.ubuntu.com/meetingology.
[15:13] <meetingology> Available commands: action commands idea info link nick
[15:13] <bdmurray> hooray for cyphermox
[15:13] <cyphermox> echo $(shuf -e barry doko bdmurray slangasek caribou infinity sil2100 robru cyphermox tdaitx xnox chiluk mwhudson rbalint)
[15:14] <cyphermox> robru caribou sil2100 bdmurray rbalint tdaitx infinity chiluk barry slangasek doko mwhudson xnox cyphermox
[15:14] <cyphermox> maybe this list is out of date.
[15:14] <sil2100> Yeah
[15:14] <bdmurray> :-(
[15:14] <cyphermox> :(
[15:14] <tdaitx> damn =(
[15:14] <cyphermox> sil2100: you go
[15:14] <sil2100> Need to find my list
[15:14] <caribou> yeah, just dropped by to say hi :)
[15:14] <sil2100> caribou: hey ;)
[15:14] <sil2100> - Bileto:
[15:14] <sil2100>   * Forced clearing the lru_cache for Bileto to enable artful
[15:14] <sil2100>   * Looked into missing GPG key errors, filled RT, helped resolve with webops
[15:14] <sil2100> - ubuntu-image:
[15:14] <sil2100>   * Helping out in debugging broken github PR test result reporting, filled RT
[15:14] <sil2100>   * Filled RT for upload rights for snap
[15:14] <sil2100>   * Prepared existing branches for merging
[15:15] <sil2100>   * Started work on the hook mechanism - planning and experimental phase
[15:15] <sil2100> - Kernel SRUs
[15:15] <cyphermox> hey caribou!
[15:15] <sil2100> - Regular SRU reviews
[15:15] <sil2100> - Cleanup on uvp-monitor test packages, reading up on xenstore - first builds
[15:15] <sil2100> - DMB meeting
[15:15] <sil2100> - walinuxagent published to -proposed, waiting for SRU validation
[15:15] <sil2100> (done)
[15:15] <bdmurray> Wrote up test case for LP: #1676547
[15:15] <bdmurray> Foundations infrastructure meeting
[15:15] <bdmurray> Discussed an approach to get Package info on ubuntu core
[15:15] <bdmurray> Started work on having Package written client side
[15:15] <bdmurray> Modified snapport to have apport.cli command not apport.apport-cli
[15:15] <bdmurray> Updated StableReleaseUpdates wiki page re: dev release being closed
[15:15] <bdmurray> added artful to errors, daisy code
[15:15] <bdmurray> added artful to lp-retracer-config, updated osageorange
[15:15] <bdmurray> updated u-r-u for artful, uploaded it
[15:15] <bdmurray> updated meta-release-development for artful, updated server
[15:15] <bdmurray> reported / investigated u-r-u bug LP: #1686117
[15:16] <bdmurray> added artful to the official bug tags
[15:16] <bdmurray> resolved issue with daisy not cleaning up core files (LP: #1686135)
[15:16] <bdmurray> submitted RT regarding updating daisy in staging w/ core file fix
[15:16] <bdmurray> tested update of daisy in staging, submitted RT for production
[15:16] <bdmurray> foundations bug triage
[15:16] <bdmurray> irc discussion re LP: #1681231 and LP: #1676547 (using -security for upgrades)
[15:16] <bdmurray> investigation into LP: #1682934 re local python3 and upgrade failure
[15:16] <bdmurray> ✔ done
[15:16] <rbalint> * Finished my core-dev application
[15:16] <rbalint> * Prepared unminimize script, under review:
[15:16] <rbalint>   https://code.launchpad.net/~rbalint/livecd-rootfs/minimize-unminimize/+merge/323157
[15:16] <rbalint> * Started checking each package with delta in main and noticed that Ubuntu history
[15:16] <rbalint>   is not available anymore in bzr: LP: #1686100
[15:16] <rbalint> * Packaging guide also lists gone lintian result server: LP: #1686096
[15:16] <rbalint> * grab-merge with also geting Debian's packaging repo/source with dgit works nicely for packages not in usd-importer yet
[15:16] <rbalint> * Sent Ubuntu deltas to Debian:
[15:16] <rbalint>   main-menu #861237
[15:17] <rbalint>   libdebian-installer #861241
[15:17] <rbalint>   cdrom-detect #861300
[15:17] <rbalint> * Proposed fix for MoM to let people find the comment box to fill
[15:17] <rbalint>   LP: #1686640
[15:17] <rbalint> * I would like to go through the page and provide Debian bug numbers for the forwarded delta or an explanation for not forwarding all of it
[15:17] <rbalint> eof
[15:17] <tdaitx> Short week: holiday on Friday
[15:17] <tdaitx> = JCK
[15:17] <tdaitx>   * Waiting on 8u131
[15:17] <tdaitx> = OpenJDK
[15:17] <tdaitx>   * No more test regressions
[15:17] <tdaitx>   * Preparing package for the security team
[15:17] <tdaitx>   * Backporting patches to OpenJDK 7
[15:17] <tdaitx> = Other
[15:17] <tdaitx>   * National strike on Friday, I will be working
[15:17] <tdaitx>   * Yet another holiday on Monday, I might decide to swap it
[15:17] <tdaitx> (done)
[15:18] <infinity>  * Archive opening for Artful:
[15:18] <infinity>    - toolchain mangling
[15:18] <infinity>    - dpkg/debhelper merges
[15:18] <infinity>    - open archive
[15:18] <infinity>    - update cdimage, debian-cd, snakefruit, etc for artful
[15:18] <infinity>    - update d-i
[15:18] <infinity>    - enable image builds
[15:18] <infinity>  * snakefruit and nusakan upgrades to xenial
[15:18] <infinity>  * investigate LP: #1682934
[15:18] <infinity>  * Misc SRU/AA bits
[15:18] <infinity>  * On vacation all next week \o/
[15:18] <infinity> (done)
[15:19] <gaughen> o/
[15:19] <sil2100> \o
[15:19] <tdaitx> o/
[15:20] <xnox> is it me?
[15:20] <tdaitx> yeah
[15:20] <cyphermox> doko isn't here?
[15:20] <gaughen> only you know if it is you, xnox
[15:20] <cyphermox> oh yeah
[15:20] <xnox> cloud images for onmetal fun
[15:20] <xnox> apt timers bug fun
[15:20] <xnox> wsl fun
[15:20] <xnox> upstart removal fun
[15:20] <xnox> open-vm-tools sru fun
[15:20] <xnox> systemd artful & sru fun
[15:20] <xnox> ...
[15:20] <tdaitx> oh, fun status, yay!
[15:20] <cyphermox> MIR:
[15:20] <cyphermox>  - review gjs and mozjs38 (both for GNOME Shell)
[15:20] <cyphermox> netplan:
[15:20] <cyphermox>  - merging and testing rharper's MTU work
[15:20] <cyphermox> grub2:
[15:20] <cyphermox>  - reworking fbx64.efi's installation process / location
[15:20] <cyphermox>  - handle upgrade / moving fbx64.efi
[15:21] <cyphermox>  - install to /EFI/BOOT too (reworking a patch from Debian)
[15:21] <cyphermox>  - ship BOOTX64.CSV (fallback instructions) to rebuild BootOrder.
[15:21] <cyphermox> d-i:
[15:21] <cyphermox>  - starting d-i merges: installation-report, tzsetup.
[15:21] <cyphermox>  - coordinating merges with rbalint.
[15:21] <cyphermox> shim / fwupdate:
[15:21] <cyphermox>  - some discussion about signtool.exe being broken (would break new shim signatures from MS)
[15:21] <cyphermox>  - coordinating fwupdate updates to Yakkety and Xenial.
[15:21] <cyphermox> (done)
[15:21] <cyphermox> that's it for the lightning rounds subtopic I forgot to start.
[15:21] <cyphermox> #subtopic AOB
[15:21] <cyphermox> err
[15:22] <cyphermox> any AOB anyway?
[15:22] <rbalint> sorry, what is AOB?
[15:22] <cyphermox> [TOPIC] AOB
[15:22] <cyphermox> any other business?
[15:23] <cyphermox> other random stuff we need to talk about
[15:23] <rbalint> oh, I would like to ask how we should coordinate visiting all the packages with delta
[15:23] <rbalint> i would happily go through each of them
[15:23] <cyphermox> what do you mean by how to coordinate?
[15:24] <rbalint> should I ping the last uploader before i check each package?
[15:24] <bdmurray> rbalint: There's an easy u-r-u bug if you want to have a look.
[15:24] <rbalint> bdmurray: ok, i'll check that too, but looking forward
[15:25] <cyphermox> my understanding was that one should ping the last uploader listed on merges.u.c and make sure before claiming a merge
[15:26] <nacc> [lurking, not on foundations, but what cyphermox said was what i was told as well]
[15:26] <cyphermox> partly because some may *look* easy, but aren't, and missing some bit in a merge because it looks like it's not needed can break things.
[15:28] <rbalint> ok, then probably i'll ping owners of a block of 20 packages then i'll process them next day as i get green light
[15:28] <rbalint> if that's ok
[15:29] <bdmurray> Can I stop reviewing Precise SRUs?
[15:29] <xnox> no
[15:29] <xnox> lol =)
[15:29] <cyphermox> rbalint: I think I can already give you green light for any d-i package I was the last to touch; but don't hesitate to ask questions if something isn't obvious
[15:29] <cyphermox> anything else?
[15:30] <rbalint> cyphermox: thanks
[15:30] <bdmurray> cyphermox: Is that X to Y upgrade verified?
[15:30] <cyphermox> bdmurray: yes.
[15:31] <cyphermox> I'll go update the bug
[15:31] <cyphermox> that's all for the meeting I guess
[15:31] <cyphermox> #endmeeting
[15:31] <meetingology> Meeting ended Thu Apr 27 15:31:19 2017 UTC.
[15:31] <meetingology> Minutes:        http://ubottu.com/meetingology/logs/ubuntu-meeting/2017/ubuntu-meeting.2017-04-27-15.13.moin.txt
[15:32] <sil2100> o/
[15:32] <sil2100> Thanks cyphermox !
[15:32] <infinity> rbalint: Some of these deltas forwarded to Debian don't make much sense.
[15:32] <rbalint> infinity: which one?
[15:33] <infinity> rbalint: Well, cdrom-detect is the first one I looked at, and you're forwarding deltas that change Debian to Ubuntu. :P
[15:33] <cyphermox> infinity: I looked too, but it's confusing, it's a patch of a patch.
[15:33] <rbalint> infinity: it is applied conditionally
[15:34] <nacc> rbalint: also, i assume you'd check any notes at https://merges.ubuntu.com/main.html or otherwise
[15:34] <infinity> rbalint: It... Is?
[15:34] <rbalint> nacc: sure
[15:35] <cyphermox> +ifeq (yes,$(shell dpkg-vendor --derives-from Ubuntu && echo yes))
[15:35] <cyphermox> +	patch -p1 < $(CURDIR)/debian/patches/ubuntu-installer.patch
[15:35] <cyphermox> +endif
[15:35] <infinity> Oh, gross.
[15:35] <cyphermox> there is that
[15:35] <cyphermox> I wonder if it's not better to just keep delta for that.
[15:35] <infinity> rbalint: I'd really rather keep deltas than introduce patch systems for native packages. :P
[15:36] <infinity> rbalint: And deltas like this are ones we need to touch regularly anyway, there's no value in pushing them back to Debian.
[15:36] <rbalint> cyphermox: i think accepting syncs is way easier than merging
[15:37] <cyphermox> rbalint: it's not "really" going to be easier for the regular "new release is named XYZ" process
[15:37] <infinity> Right, for a new release, I'm not going to push to Debian and sync.
[15:37] <infinity> That's a delay I can't be bothered with.
[15:37] <cyphermox> you still need to touch the package, so that you're adding a step
[15:37] <infinity> So we'll get a delta every 6 months on that hunk.
[15:38] <infinity> rbalint: So, perhaps this answers your other question, yes you should ask the "owner" of the delta before forwarding it. :P
[15:39] <rbalint> infinity: ok
[15:40] <infinity> rbalint: In some cases, it really is less hassle to maintain a small delta than to shift to maintaining a dpkg-vendor delta in Debian via bug reports.
[15:40] <infinity> rbalint: In other cases, we intentionally carry a delta to gate flow into Ubuntu (dpkg and debhelper come to mind here, I could reduce both to 0, and I won't)
[15:40] <cyphermox> infinity: not really maintaining via bug reports though
[15:41] <infinity> cyphermox: For d-i, sure, I can commit.  That's not true of everything.
[15:41] <cyphermox> I think we (you, me, maybe rbalit too)
[15:41] <cyphermox> yeah, for d-i
[15:41] <infinity> cyphermox: I can upload everything because DD, but Debian doesn't have the same weak maintainership policy and people aren't generally happy with you uploading their stuff. :P
[15:41] <cyphermox> definitely not true for everything, but it's not as much as issue for non d-i.
[15:41] <cyphermox> no, I know that
[15:42] <cyphermox> I meant specifically for d-i where we can already commit to the tree; and it's more likely to get such branding issues
[15:42] <cyphermox> I don't expect to see much branding or "totally not applicable to debian" changes in say, miniupnpc, to pick something randomly.
[15:43] <rbalint> infinity: not pushing changing Ubuntu codenames makes sense
[15:44] <rbalint> infinity: i think if Debian's maintainter accepts the vendor-based changes they can be offered
[15:44] <cyphermox> rbalint: it's also about using a custom patch system on a native package... which *is* quite yucky
[15:44] <rbalint> rbalint: i gladly merged raspbian changes
[15:44] <infinity> Very.
[15:45] <infinity> I think adding a patch system to d-i packages is a non-starter for me.
[15:45] <cyphermox> perhaps for branding (not release names), it could be argued that proper application of db_subst in the templates could do a better job.
[15:45] <rbalint> infinity: ok
[15:46] <rbalint> .in file processing would be a no-go, too?
[15:46] <infinity> It's a lot of added complexity to a package you're not maintaining. :P
[15:46] <infinity> rbalint: I understand that you find it's less hassle for you to sync instead of merge, but if no one's asking you to do either of those things...
[15:59] <rbalint> ok, won't push for pushing