[07:03] <mwhudson> is there some deep reason why gcc-multilib and the cross compiler packages can't be coinstalled?
[07:23] <pitti> Good morning
[08:04] <dholbach> good morning
[08:04] <zyga> good morning
[08:11] <sladen> guten morgen
=== DrKranz is now known as DktrKranz
=== nudtrobert1 is now known as nudtrobert
[10:39] <Trevinho> doko: unity doesn't link on arm64 due to a segfault... I've no clue how to debug thatù https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-011/+build/8215078
[11:26] <doko> Trevinho, loking
[11:36] <Trevinho> doko: thanks
=== _salem is now known as salem_
[12:23] <Trevinho> doko: got anything?
[12:29] <doko> Trevinho, will you ask every 30min?
[12:31] <Trevinho> doko: I just asked once, I didn't want to bother you...
=== hikiko-lpt is now known as hikiko|ln
[12:59] <doko> Trevinho, does the package build on wily?
[13:02] <doko> Trevinho, I can't reproduce this. already fails earlier with http://paste.ubuntu.com/12998753/
[13:02] <doko> so please fix this, then I can look again
[13:03] <sitter> balloons: ping. about the kubuntu ci session. Id' be entirely fine if that slot were to be halved... just in case another session gets proposed ;). At akademy shadeslayer and I did a talk about our various CI systems and that fit nicely in a 30 minute slot
[13:18] <pitti> cjwatson, wgrant: OOI, is it planned to allow github issues as upstream bug URLs in LP?
=== hikiko|ln is now known as hikiko
[13:47] <pitti> cjwatson: do you plan to do https://merges.ubuntu.com/f/finish-install/, or want me to steal it?
[13:47] <pitti> cjwatson: it's important for https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=802187
[13:47] <ubottu> Debian bug 802187 in finish-install "debian-installer: On first boot after install init hangs due to missing symlink "/etc/mtab to /proc/self/mounts"" [Serious,Fixed]
[13:51] <rbasak> pitti: no plans for a migration path for the mtab symlink?
[13:51] <pitti> rbasak: we already have that
[13:52] <rbasak> Oh, OK>
[13:52] <pitti> rbasak: but we should stop writing /etc/mtab in installers as a file, and set it up correctly right away
[13:52] <pitti> rbasak: it's turned into a symlink on every boot
[13:52] <pitti> this is to find the installer bugs
[13:52] <pitti> rbasak: I just mailed u-devel@ with some explanation
[13:53] <rbasak> Oh sorry. I was responding to the email but missed your mention of migration.
[14:03] <Trevinho> doko: in wily it worked...
[14:04] <Trevinho> doko: that's already fixed..
[14:04] <Trevinho> You need the 7.4 bump branch
[14:06] <Trevinho> doko: https://code.launchpad.net/~unity-team/unity/unity-7.4.0-bump
[14:07] <doko> looking
[14:18] <doko> Trevinho, still fails. I can't install dependencies on the porter box, which are not in the archive. will have to wait until I can setup one, so tomorrow
[14:19] <cjwatson> pitti: github linking is on our backlog though it's not trivial
[14:19] <cjwatson> pitti: I can do finish-install, sure
[14:24] <cjwatson> ogra_: is it OK if I upload your livecd-rootfs changes?
[14:24] <ogra_> cjwatson, perfectly fine ...
[14:24] <Trevinho> doko: Mh ok... If you want, all the packages needed are in silo 11 (ppa)
[14:24] <cjwatson> thanks
[14:24] <Trevinho> You can use those deb sources for  building it
[14:27] <cjwatson> pitti: done
[14:38] <pitti> cjwatson: oh, that was fast, thanks!
[14:54] <sil2100> cjwatson: hey! Could we request an LP translation export for ubuntu-rtm/15.04? It's usually taking around 5 minutes so it shouldn't eat up much resources I suppose
[14:56] <cjwatson> sil2100: requested
[14:56] <sil2100> cjwatson: thank you!
[14:56] <cjwatson> sil2100: (see #webops)
[15:06] <Chir0n> hi there! keyserver.ubuntu.com seems to be down, known fact?
[15:06] <Chir0n> (not sure what a good channel is to ask about it)
[15:07] <cjwatson> Chir0n: I see it on our internal outage channel so our sysadmins know about it
[15:07] <cjwatson> via nagios
[15:08] <Chir0n> cool, thanks!
[15:20] <tkamppeter> cjwatson, hi
[15:21] <cjwatson> tkamppeter: hi, what's up?
[15:22] <seb128> cjwatson, unping, I'm on it, Till wants pysmbc removed from the sync blacklist
[15:22] <cjwatson> ok
[15:23] <tkamppeter> cjwatson, I want to have our python-smbc replaced by Debian's pysmbc. I do not see any reason why we are going separate here. What I need from you for that is to remove pysmbc from http://people.canonical.com/~ubuntu-archive/sync-blacklist.txt.
[15:25] <tkamppeter> cjwatson, seems that seb128 did it already, but please check whether he did it correctly.
[15:26] <seb128> well I edited the vcs
[15:26] <seb128> unsure when the cron job is running
[15:27] <cjwatson> sil2100: should be done for you now
[15:27] <cjwatson> seb128: it's every five minutes
[15:27] <sil2100> \o/ thanks!
[15:27] <tkamppeter> seb128, now it worked. I have synced now.
[15:27] <seb128> cjwatson, thanks
[15:28] <tkamppeter> cjwatson, seb128, thanks.
[15:33] <Chir0n> looks like keyserver is back!
[15:34] <Chir0n> someone tripped over a cable? :)
[15:34] <cjwatson> unlikely
[15:34] <cjwatson> something to do with a package upgrade but I don't know exactly what
[15:43] <slangasek> cyphermox, cjwatson: it's possible for grub to chainload to another instance of grub.efi, right?  if so what's the syntax?
[15:43] <slangasek> cyphermox, cjwatson: bug #1510120; PXE-booted GRUB is trying to find kernels on LVM, and I'd prefer that we chainload to the GRUB on disk rather than adding a pile of disk modules to grubnetx64.efi.signed if we can
[15:43] <ubottu> bug 1510120 in MAAS "UEFI deployment broken in MAAS 1.9" [Critical,Triaged] https://launchpad.net/bugs/1510120
[15:45] <cjwatson> slangasek: chainloader <filename>
[15:45] <cjwatson> slangasek: but I don't see how that would work with secure boot
[15:46] <blake_r> cjwatson: the grub on disk should be signed as well
[15:46] <slangasek> cjwatson: oh?  if both files are signed, it still won't work?
[15:46] <cjwatson> blake_r,slangasek: well that would be great if chainloader had any sig verification support
[15:47] <slangasek> oh, well then
[15:47] <slangasek> is that why we are failing to chainload Windows?
[15:47] <cjwatson> hm
[15:47] <cjwatson> and yet we put chainloader in the signed image
[15:47] <cjwatson> um, I'm too far out of context on this, you might have to try it
[15:48] <slangasek> I thought chainloader in a UEFI context would just tell UEFI "execute this", and the "execute this" API does the signature checking
[15:48]  * cyphermox looks
[15:49] <slangasek> blake_r: so yes, since you've already found the ESP you can try chainloader .../shimx64.efi instead of loading the config
[15:49] <blake_r> slangasek: would I chainload the shim or grub? as the shim was already loaded over network
[15:49] <sil2100> pitti: hey! Could you manually run langpack-o-matic on ubuntu-rtm/15.04 for us? :)
[15:50] <slangasek> blake_r: shim.  a chainload goes back to the firmware (I believe; and if it doesn't it needs to), which means the firmware itself needs to be able to verify the sig
[15:50] <blake_r> slangasek: okay  I will try it
[15:52] <cjwatson> slangasek: if that is so then all should be well
[15:53] <cyphermox> looks to me like it calls LoadImage, which should return a SECURITY_VIOLATION if the signature is bad
[15:54]  * slangasek nods
[15:59] <blake_r> slangasek: worked
[15:59] <blake_r> slangasek: thanks, easy fix
[16:00] <slangasek> blake_r: hurray
=== davidcalle is now known as davidcalle|afk
[17:37] <ogra_> mdeslaur, wow, i'm surprised you also took MODE="0666" for the steam fix ... instead of just using uaccess everywhere
[17:37] <mdeslaur> hrm, I guess I could have done that...but I'm not sure what process actually updates the firmware in those controllers
[17:38]  * ogra_ finds 0666 always a little itchy 
[17:38] <ogra_> but yeah, if the frimware updater runs as different user etc ...
[17:38]  * mdeslaur nods
[17:58] <dobey> fw updater?
[18:14] <pitti> sil2100: uh, did the cronjob not run?
[18:15] <pitti> sil2100: https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/stable-phone-overlay has packages from Tuesday, so it did run; so you need the export from today?
[18:15] <pitti> sil2100: running
[18:16] <sil2100> pitti: thanks! It ran, yes, but we requested another export today to get the latest stuff after feature freeze
[18:26] <sil2100> doko, cyphermox, slangasek: pushed a batch of new rebuilds for the ocaml transition - some are unresolvable for now, like xstrp4 (I mentioned that yesterday) and now also lablgtk-extras seems to be requiring the new camlp4 and ocaml from Debian to build properly
[18:27] <sil2100> Give me a sign when we could merge those from Debian
[18:28] <sil2100> There also seems to be a number of false positives on the transition page, at least the uninstallabilities are not reproducible on my chroots
[18:29] <slangasek> sil2100: I don't understand - give you a sign when?  why wouldn't they get merged now?
[18:29] <slangasek> (I haven't been looking at ocaml at all, I have no context here)
[18:29] <sil2100> slangasek: I poked doko earlier about that and he mentioned we should wait with that as Debian is peraring a new ocaml version anyway
[18:30] <sil2100> But I'm worried we'll have to rebuild once again when that happens
[18:31] <slangasek> sil2100: does the update from ocaml 4.01.0 to 4.02.3 require rebuilds?
[18:33] <doko> slangasek, sil2100: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=789133   it's already ongoing
[18:33] <ubottu> Debian bug 789133 in release.debian.org "transition: ocaml 4.02.3" [Normal,Open]
[18:33] <slangasek> ok
[18:33] <sil2100> I suppose it will, as the provides package would change
[18:34] <slangasek> sil2100: so that's the version already in unstable; if that's the version needed for unblocking parts of the transition you're looking at, we should merge that asap so we're doing 1 transition instead of 2
[18:34] <sil2100> And some packages already require ocaml 4.02.3 to work and build
[18:34] <slangasek> right.  as long as jumping directly to 4.02.3 doesn't introduce problems because of versioned build-dep loops, we should jump straight to that
[18:46] <pitti> sil2100: done, in the ppa
[18:46]  * pitti waves good night
[18:56] <Saviq> oh, do-release-upgrade to xenial did not snapshot my btrfs root, isn't that what happened on btrfs usually?
[19:01] <hallyn> so, qemu failed to build in power8;  while i was still trying elsewhere to figure out why, a rebuild seems to have been triggered and it passed.  What would cause an automatic rebuild like that?
[19:04] <sil2100> slangasek: ok, I'll prepare a merge, I'll need a sponsor once I'm done :)
[19:05] <slangasek> sil2100: good news I'm piloting tomorrow
[19:05] <sil2100> Excellent
[19:41] <cjwatson> hallyn: nothing automatic, but quite a few people have access to retry builds manually
=== davidcalle|afk is now known as davidcalle
[20:31] <sil2100> slangasek: if anything, I filled in #1511517 - will wait with further ocaml transitioning for now until this is landed, switching back to some touch work for the rest of today
[20:31] <sil2100> LP: #1511517 of course
[20:31] <ubottu> Launchpad bug 1511517 in ocaml (Ubuntu) "Please merge ocaml 4.02.3-5 (main) from Debian unstable" [Wishlist,In progress] https://launchpad.net/bugs/1511517
[20:31] <sil2100> slangasek: I hope I did the merge right
[20:31] <sil2100> pitti: thanks! And good night to you too :)
=== cmagina_ is now known as cmagina
=== salem_ is now known as _salem
[22:27] <slangasek> sil2100: ocaml uploaded
[23:34] <sil2100> slangasek: \o/ thanks!
[23:44] <slangasek> man, why does gnome-pkg-tools still Recommend svn-buildpackage in Debian