UbuntuIRC / 2020 /06 /23 /#ubuntu-devel.txt
niansa
Initial commit
4aa5fce
=== JanC_ is now known as JanC
[06:46] <bryce> teward, https://code.launchpad.net/~bryce/ubuntu/+source/nginx/+git/nginx/+merge/386231
[08:37] <Laney> seb128: LocutusOfBorg: what's up with the sane stuff? libsane1 needs promoting after all or?
[08:42] <seb128> Laney, yes, LocutusOfBorg changed his mind, I will look at that after the meeting I'm currently in
[08:43] <Laney> right, I looked at the package and it seems that libsane is the one that should be demoted eventually
[08:43] <Laney> thanks!
[08:46] <seb128> right, libsane1 is the one to promote now
[08:54] <LocutusOfBorg> yes, sorry for saying the exact opposite :/
[08:55] <LocutusOfBorg> we transitioned that library back and forth around 3-4 times in the last 2 years, and no ABI changes at all
[08:55] <RAOF> Didn't I already promote that?
[08:55] <Laney> I think there was some confusion :P
[08:55] <LocutusOfBorg> this sucks a lot, because people kept syncing it over and over from experimental, debian changed its mind and reverted that change, but we had already transitioned
[08:59] <seb128> RAOF, the binary is still in universe, if you want to try again please do, otherwise I've a look after that meeting
=== ricab__ is now known as ricab
[09:26] <LocutusOfBorg> also RAOF please move src:libnma to main, the binaries are already in main :) https://people.canonical.com/~ubuntu-archive/component-mismatches.html
[09:27] <LocutusOfBorg> and I think we can also demote libsane to universe, but I'm not quite sure about that
[09:39] <Laney> wait for component-mismatches to say
[10:12] <seb128> xnox, could you look at sponsoring the update on bug #1882185 ? Olivier is out this week but he said that's going to be needed for the new firefox that is due for next week
[10:12] <ubottu> bug 1882185 in nodejs (Ubuntu) "Firefox 78 requires nodejs >= 10.21" [High,In progress] https://launchpad.net/bugs/1882185
[10:17] <seb128> hum, security updates bypass the autopkgtest infra? :(
[10:18] <LocutusOfBorg> seb128, problem is that nodejs regressed a lot of stuff :/
[10:18] <LocutusOfBorg> but meh, I can also have a look if x nox is away
[10:18] <seb128> LocutusOfBorg, how do you know if that version hasn't landed yet?
[10:18] <seb128> LocutusOfBorg, also we are going to need to update firefox one way or another
[10:18] <seb128> LocutusOfBorg, thanks
[10:22] <LocutusOfBorg> https://packages.qa.debian.org/nodejs
[10:22] <LocutusOfBorg> seb128, the debian tracker shows them, and nodejs are pretty much packages in sync...
[10:22] <LocutusOfBorg> in any case, lets upload and see what happens
[10:23] <seb128> LocutusOfBorg, thx
[10:38] <mwhudson> is ubuntuwire down?
[10:40] <seb128> mwhudson, seems so, probably something to mention to IS?
[10:40] <mwhudson> i don't think we run it though
[10:41] <seb128> wgrant might know
[10:42] <mwhudson> yeah he is on https://launchpad.net/~ubuntuwire-sysadmins/+members#active
[10:42] <seb128> there is a #ubuntuwire also
[10:42] <wgrant> Looking
[10:46] <wgrant> seb128, mwhudson: Is back.
[10:46] <seb128> wgrant, thx!
[10:47] <mwhudson> wgrant: thanks
=== alan_g_ is now known as alan_g
[12:18] <LocutusOfBorg> wgrant, can you please help understanding this setarch failure? https://launchpad.net/ubuntu/+source/util-linux/2.35.2-4ubuntu1/+build/19497777
[12:20] <wgrant> LocutusOfBorg: It's not one I know about, but vorlon at least glanced at it a few weeks back.,
[12:26] <LocutusOfBorg> I mean, wgrant this command setarch riscv64 -v --uname-2.6 seems to give segfault, but I don't know if qemu is to blame or not
[12:26] <LocutusOfBorg> do you have a possibility to launch that command?
[12:26] <LocutusOfBorg> and also setarch from the old util-linux in release has the same segfault, just to be sure... the regression might be in something else, kernel maybe?
[12:33] <wgrant> LocutusOfBorg: Oh really, pretty sure that used to work
[12:33] <wgrant> Let me see
[12:33] <wgrant> I just put away my board because my cat was trying to bite the fan
[12:33] <wgrant> She'll just have to deal with it
[12:34] <wgrant> There was an issue with related code that assumed a glibc thing was static, but it become non-static and broke on like m68k and riscv64
[12:34] <wgrant> I wonder if this is related.
[12:34] <LocutusOfBorg> the code of setarch didn't change at all, so something else is going under the hood, but trying to gdb it gives lots of uninplemented stuff
[12:34] <LocutusOfBorg> mmm interesting
[12:37] <LocutusOfBorg> linux changed from 5.3 to 5.4, glibc from 2.31-0ubuntu7 to ubuntu10
[12:38] <wgrant> LocutusOfBorg: Where did you see that the old one has the same segfault?
[12:38] <wgrant> Oh, building the old version in a PPA or something?
[12:39] <LocutusOfBorg> nope, ppa is sad
[12:39] <LocutusOfBorg> pbuilder chroot local
[12:40] <LocutusOfBorg> based on this build, https://launchpad.net/ubuntu/+source/util-linux/2.35.1-5ubuntu2/+build/19330488 changes in toolchain are not that many
[12:50] <LocutusOfBorg> https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/4114/+build/19498613
[12:50] <LocutusOfBorg> this is the focal version just no change rebuilt in bileto
[14:03] <LocutusOfBorg> old one still builds with focal
[14:03] <LocutusOfBorg> lets try the new one with focal https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/4114/+build/19499332
[14:31] <seb128> LocutusOfBorg, argyll/riscv seems not happy (ftbfs)
[14:57] <LocutusOfBorg> seb128, nack, not a regression
[14:57] <seb128> LocutusOfBorg, I didn't speak of regression
[14:57] <seb128> https://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses_by_team.html#desktop-packages just state
[14:57] <seb128> colord
[14:57] <seb128> Unsatisfiable depends:
[14:57] <seb128> argyll: riscv64
[14:57] <LocutusOfBorg> I wanted to no change rebuild just because the riscv64 was not retryable
[14:57] <seb128> I see, the changelog was confusing
[14:58] <seb128> I though you meant it would build now :)
[14:58] <LocutusOfBorg> seb128, yes, but meh
[14:58] <LocutusOfBorg> I guess britney will consider it and let it migrate anyway
[14:58] <LocutusOfBorg> it is not installable on riscv64 but also on release pocket
[15:00] <LocutusOfBorg> what is the workflow for a library that was in main, not in universe, and we want it in main again?
[15:00] <LocutusOfBorg> https://launchpad.net/ubuntu/+source/xxhash
[15:01] <LocutusOfBorg> (new rsync is trying to use it in proposed)
[15:03] <seb128> LocutusOfBorg, read https://irclogs.ubuntu.com/2020/06/23/%23ubuntu-meeting.html
[15:10] <LocutusOfBorg> ta
=== grumble is now known as rawr
[16:10] <teward> bryce: thanks for the link. cpaelzer brings up some good points, but I think we need to check to see the difference between Debian and us, some things may need poked up there for them failing for things.
[16:25] <LocutusOfBorg> Laney, looks like component is now saying it, thanks! libsane sane-backends
[16:25] <LocutusOfBorg> seb128, ^^ :)
[17:07] <Odd_Bloke> slyon: bdmurray: Have you seen https://bugs.launchpad.net/ubuntu/+source/apport/+bug/1884221 ? (It appears to be fixed in lp:apport.)
[17:07] <ubottu> Launchpad bug 1884221 in apport (Ubuntu) "`ubuntu-bug` fails with "UnboundLocalError: local variable 'project' referenced before assignment"" [Undecided,Confirmed]
[17:09] <bdmurray> Odd_Bloke: I've seen it but haven't dug into it yet
[17:10] <Odd_Bloke> bdmurray: Want me to take a look?
[17:12] <bdmurray> Odd_Bloke: Sure if you are interested.
[17:13] <Odd_Bloke> bdmurray: What's the appropriate way to propose the change?
[17:16] <bdmurray> Odd_Bloke: an MP against the groovy branch - not upstream
[17:17] <Odd_Bloke> bdmurray: Including changelog entry?
[17:18] <bdmurray> Odd_Bloke: yes please
[17:21] <rafaeldtinoco> ahasenack: https://code.launchpad.net/~rafaeldtinoco/ubuntu/+source/autofs/+git/autofs/+merge/386267
[17:21] <rafaeldtinoco> if you have time for a quick +1
[17:21] <rafaeldtinoco> its the same as the previous, without the & quoting as well
[17:21] <rafaeldtinoco> I'll keep the & for the SRUs since its not backed by an upstream change and we are are fixing the $ behavior only
=== ijohnson is now known as ijohnson|lunch
[17:25] <ahasenack> rafaeldtinoco: ok
[17:27] <Odd_Bloke> bdmurray: https://code.launchpad.net/~oddbloke/apport/lp1884221/+merge/386269
[17:32] <bdmurray> Odd_Bloke: thanks!
=== ijohnson|lunch is now known as ijohnson
[18:31] <bdmurray> jibel / xnox: can the SRU information in bug 1875045 be updated?
[18:31] <ubottu> bug 1875045 in ubiquity (Ubuntu) "Ubiquity 20.04 exports existing ZFS pools" [Low,Fix released] https://launchpad.net/bugs/1875045
[18:31] <bdmurray> and bug 1880869
[18:31] <ubottu> bug 1880869 in ubiquity (Ubuntu) "Use persistent device name for vdevs" [High,Fix released] https://launchpad.net/bugs/1880869
[18:46] <xnox> jibel: i did ask about it before. It's not blocking testing ubiquity for the point release.
[18:46] <xnox> jibel: shall i drop the zfs backports, and reupload SRU without them?
[18:57] <jibel> xnox, do not drop the backport, I'll update the bugs and do the verification this week
[19:01] <xnox> tah
[20:33] <rafaeldtinoco> are there any plans for ifupdown deprecation ?
[20:34] <ogra> didnt that happen in 18.04 ?
[20:43] <rafaeldtinoco> I meant removal, sorry
[20:52] <rafaeldtinoco> ddstreet: ^ this was for our merge-review discussion
[20:52] <rafaeldtinoco> I think the sync approach and "let it go" is the best way
[20:52] <rafaeldtinoco> will seek that