| === gurupras- is now known as guruprasad | ||
| === tribaal_ is now known as tribaal | ||
| utkarsh2102 | hey, livecd-rootfs' autopkgtest is failing for s390x in Jammy because it can't open http://people.canonical.com/~ubuntu-archive/seeds/ubuntu.jammy/STRUCTURE!?!?!?!? | 09:23 |
|---|---|---|
| utkarsh2102 | logs - https://autopkgtest.ubuntu.com/results/autopkgtest-jammy/jammy/s390x/l/livecd-rootfs/20240812_065114_ae836@/log.gz | 09:23 |
| utkarsh2102 | any idea what's wrong? :/ | 09:24 |
| xnox | hmmm no ahasenack | 09:39 |
| xnox | cc ubuntu-server / ahasenack https://bugs.launchpad.net/ubuntu/+source/multipath-tools/+bug/2076598 | 09:42 |
| -ubottu:#ubuntu-devel- Launchpad bug 2076598 in multipath-tools (Ubuntu) "friendly-names by default" [Undecided, New] | 09:42 | |
| utkarsh2102 | xnox: ohaiiii! | 09:55 |
| utkarsh2102 | long time, no see :) | 09:56 |
| utkarsh2102 | also, xnox, ahasenack: Debian didn't want to do that, I had opened an MP and they were like no, not this please. :P | 09:56 |
| Rhonda | utkarsh2102: Sometimes people change their minds on things. I guess asking again after five years might be okay. The bug also mentions something about that debian-installer seems to use it anyways (though haven't checked that or whether it works or not like zeha commented) | 10:14 |
| utkarsh2102 | https://salsa.debian.org/linux-blocks-team/multipath-tools/-/merge_requests/4 -> this was the MR 3 years ago | 10:16 |
| -ubottu:#ubuntu-devel- Merge 4 in linux-blocks-team/multipath-tools "Install friendly names multipath.conf by default" [Closed] | 10:16 | |
| Rhonda | Oh. Just noticed that zeha is also on the packaging team, so saying "Debian" didn't want to do that might not be completely true, it was one person of the packaging team, another of that team didn't object to it. | 10:17 |
| Rhonda | So sometimes it makes sense in such cases to contact that person directly and ask for support, or what their opinion is after the time. And yes, only three years, which still is quite some time. :) | 10:18 |
| Rhonda | Also, hi - hope you made it home safely. :) | 10:20 |
| utkarsh2102 | hiiii, yes, I did. Hope you did too? :) | 10:26 |
| rbasak | @pilot in | 11:10 |
| === ChanServ changed the topic of #ubuntu-devel to: Archive: open | Devel of Ubuntu (not support) | Build failures: http://qa.ubuntuwire.com/ftbfs/ | #ubuntu for support and discussion of Focal-Noble | Patch Pilots: rbasak | ||
| bluca | Hi rbasak, any chance you could help with sponsoring https://bugs.launchpad.net/ubuntu/+source/archlinux-keyring/+bug/2076416 please? Thanks! | 11:26 |
| -ubottu:#ubuntu-devel- Launchpad bug 2076416 in archlinux-keyring (Ubuntu Noble) "[SRU] Upload latest archlinux-keyring from oracular to noble-proposed" [Undecided, Confirmed] | 11:26 | |
| === cimvwrccjftwvvha is now known as pacjairummlhenxv | ||
| === pacjairummlhenxv is now known as georgiag | ||
| rbasak | bluca: o/ why isn't this and distribution-gpg-keys just part of one big source package? | 12:20 |
| rbasak | One source package per outside distribution isn't going to scale for sponsors' time. | 12:20 |
| bluca | the latter is for RPM distros | 12:28 |
| bluca | they are just handled separately upstream | 12:28 |
| bluca | just like debian-keyring and ubuntu-keyring as separate | 12:29 |
| rbasak | I'd like to see explicit and deliberate consensus amongst Ubuntu developers that this is an appropriate architecture first please. | 12:29 |
| rbasak | I commented on both bugs. | 12:29 |
| bluca | fortunately it's not one package per distro, the RPM folks are quite good at collating in the same project | 12:30 |
| bluca | so it's one for those, plus one each for debian, ubuntu and arch | 12:30 |
| rbasak | How about a single trusted key for an upstream project that maintains dynamic lists? | 12:31 |
| rbasak | Users would choose just to opt in to that, and then you wouldn't need to push SRUs repeatedly. | 12:31 |
| rbasak | Getting it from the "trusted" Ubuntu archive would add no value, since Ubuntu developers wouldn't be verifying anything anyway. | 12:31 |
| bluca | are you including the ubuntu keys in that? so, drop ubuntu-archive-keyring too? | 12:32 |
| rbasak | No. Debian and Ubuntu are different for obvious reasons. | 12:32 |
| bluca | the value is not in developers verifying, as it's unpractical to verify manually a list of keys, but it's in having a verified channel that doesn't require out-of-band transport - ie, the repository is signed, so if I trust the repository, I can trust the content too | 12:33 |
| rbasak | From the perspective of what's in the archive, anyway. | 12:33 |
| rbasak | I'm not bothered by what mkosi chooses to do - it's not the recommended way of creating Ubuntu images anyway (though of course it's Free Software and people are welcome to do what they like, including to experiment with other third party mechanisms) | 12:34 |
| bluca | sorry I do not follow, why are they different? if you want to have a "single trusted key", then it follows that ubuntu and debian keyrings need to be dropped too in favour of that - otherwise it's 3 trusted keys? | 12:35 |
| rbasak | Debian and Ubuntu keyring packages can be updated at Debian and Ubuntu's pace. They are used extensively in Debian and Ubuntu's own infrastructure. | 12:35 |
| bluca | mkosi is just one consumer of pacman, zypper and dnf - they are the actual consumers of these keyrings, and one can use them without any involvement of mkosi | 12:35 |
| bluca | you cannot update debian's keyring from ubuntu, only debian can do that | 12:37 |
| bluca | it's not any different than the other two external keyrings | 12:37 |
| rbasak | it's in having a verified channel that doesn't require out-of-band transport> isn't mkosi useless without out-of-band transport to the external distribution's repositories anyway? | 12:38 |
| bluca | no, the keyring is used to verify that | 12:39 |
| rbasak | Anyway I don't see what I can add in this conversation. It's not my sole decision or anything. Achieve consensus amongst Ubuntu developers, please. | 12:39 |
| bluca | sorry I do not follow, what do you mean by "consensus among Ubuntu developers"? | 12:39 |
| rbasak | That's how we make decisions here. | 12:39 |
| bluca | sure, but my question is, what does that entails in practice? | 12:40 |
| rbasak | Maybe start with a thread on ubuntu-devel@ explaining what you're trying to achieve and why your proposed architecture is necessary. | 12:41 |
| bluca | I'm not comfortable with the "your proposed architecture" definition - these are just upstream packages as the relevant metadata shows, I didn't architect anything | 12:42 |
| bluca | all of these - debian-keyring, archlinux-keyring, distribution-gpg-keys - are already present and shipping in both Debian and Ubuntu | 12:43 |
| bluca | the latter is new in Oracular, the other 2 already ship in releases | 12:44 |
| rbasak | You can update them via Debian if you like, and Ubuntu will autosync. But by depending on stable releases being continuously updated, your architecture is just a proposal, and Ubuntu (and any other distros) should be part of the decision if that's the architecture you're depending on. | 12:47 |
| bluca | I already updated them in Debian, as I am respectively the sponsor and maintainer of those | 12:52 |
| bluca | once again, please refrain from using the "your architecture" definition, as it's not "mine" in any sense of the word, these are long-standing, independent and pre-existing upstream projects, and I am not part of any of them - thank you | 12:54 |
| rbasak | Sorry, I thought it was "proposed" that you were objecting to. I meant you as in plural, including both you and upstream, since you are both upstream from Ubuntu's perspective. But I will disambiguate as you prefer. | 12:57 |
| bluca | no problem, thank you | 13:00 |
| === Perflosopher1 is now known as Perflosopher | ||
| jbicha | ginggs: could you update the package for https://tracker.debian.org/pkg/mobile-broadband-provider-info ? | 15:07 |
| ginggs | jbicha: maybe, upstream switched build system, so packaging needs updating | 15:12 |
| ginggs | if it wasn't for that, it would uploaded a while back | 15:13 |
| jbicha | ginggs: it looks like adding Build-Depends: meson is enough | 15:21 |
| ginggs | jbicha: thanks, i'll upload in a bit | 16:14 |
| ahasenack | sgmoore: hi, around still? | 17:42 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!