[01:28] -queuebot:#ubuntu-release- New binary: visp [riscv64] (hirsute-proposed/universe) [3.4.0-1] (no packageset)
[03:05] <mwhudson> vorlon: what's the process for getting a package (libgit2) built on i386?
[03:35] <xnox> https://wiki.ubuntu.com/i386 => there is modifying dependencies & transitioning abis
[03:35] <xnox> mwhudson:  i guess this is modifying dependencies? hence you need to add source package to i386 seed in git, and ask AA to rerun update?
[05:53] <xnox> mwhudson:  rmfile, truncate file, setup loopdevice, mkfs -L scratch loopdevice => filesystem with label scratch present, would you like to continue?
[05:53] <xnox> mwhudson:  i love that on a brand new file, one has to clean it first =)))))
[07:57] -queuebot:#ubuntu-release- Unapproved: libvirt (focal-proposed/main) [6.0.0-0ubuntu8.7 => 6.0.0-0ubuntu8.8] (ubuntu-server, virt)
[07:57] -queuebot:#ubuntu-release- Unapproved: libvirt (groovy-proposed/main) [6.6.0-1ubuntu3.3 => 6.6.0-1ubuntu3.4] (ubuntu-server, virt)
=== acheronuk is now known as RikMills
[08:40] <doko> Laney: update-motd now fails consistently on all archs. is that file permission error for debian/tests? make the package a v3 package and change the permissions in the source?
[09:01] <Laney> doko: I'll look, I was trying to get results in bileto but it never ran there so I gave up and copied it to get runs in the archive
[09:04] <doko> LocutusOfBorg: visp ftbfs
[09:06] <Laney> yeah I guess it lost the +x when uploading because of 1.0
[09:06] <Laney> I ran it from the source directory rather than the .dsc when testing locally I guess, that's why it worked
[09:08] <Laney> wonder if it should be native or quilt
[09:09] <sigv> if that's a direct question of the difference: native when debian/ is in source respository/tarball directly. quilt when debian/ is added from a diff/patch on top of source tarball.
[09:10] <sigv> without going into further details, that's a way to summarize the high level.
[09:11] <sigv> impacts expectations about how updates to source versions are interpreted, and so on, but at its core it's about whether a package is built directly from original source (native) or packaging is external (quilt).
[09:11] <sigv> feels like you should know that however. :/
[09:11] <Laney> haha
[09:12] <Laney> thanks sigv, no I was more asking which one would be right to use here
[09:12] <sigv> well, the core point then still stands: does debian/ live in the source directly and packages get built from the tarballs without patching, or are custom diffs applied?
[09:14] <sigv> because you don't expect debian/patches in a `native`. that is the main development workflow difference that guides that decision...
[09:15] <Laney> The other one is that native packages always have their entire source uploaded; that's why ubuntu-wallpapers isn't one for example
[09:15] <sigv> binary/large blobs can be a pain there, makes sense.
[09:16] <LocutusOfBorg> doko, yes I saw it
[09:16] <LocutusOfBorg> will have a lokk
[09:16] <Laney> But tools these days can handle generating orig tarballs anyway, excluding the debian/ directory, so they can live together just fine in non native packages
[09:18] <sigv> if you have a native, then you are expecting all patches to be applied on the project directly as well, and can't just plonk in a debian/patches without going through upstream.
[09:19] <Laney> It does have some random changes in the diff outside of Debian patches
[09:19] <Laney> I'm going to pick native :-)
[09:19] <Laney> thanks for the discussion
[09:19] <sigv> native is a good choice, if you can push the debian/ to the source repo and maintain patches inside that source repo, yeah.
[09:19]  * sigv wishes luck
[10:09] -queuebot:#ubuntu-release- Unapproved: libvirt (bionic-proposed/main) [4.0.0-1ubuntu8.18 => 4.0.0-1ubuntu8.19] (ubuntu-server, virt)
[10:09] -queuebot:#ubuntu-release- Unapproved: libvirt (xenial-proposed/main) [1.3.1-1ubuntu10.30 => 1.3.1-1ubuntu10.31] (ubuntu-server, virt)
[11:10] <xnox> Laney:  removing tag off grub2, and it should migrate fine this time, because it is built with i386 in hirsute now.
[11:11] <Laney> best of luck little grubs
[11:13] <Laney> cjwatson: I think ubuntu-archive-robot's email address should be updatable when you've got a moment to do that. Looks like the list works
[11:33] <paride> o/
[11:34] <paride> the last two hirsute-live-server-amd64 ISOs install unbootable systems when installing in UEFI mode
[11:35] <xnox> paride:  perfect.
[11:36] <xnox> paride:  cd-boot-images using the new grub with lots of security patches landed two days ago.
[11:36] <cjwatson> Laney: are you able to accept my list subscription?
[11:36] <xnox> paride:  but not yet grub. So iso boots, but the installed system doesn't?
[11:36] <paride> xnox, yes exactly
[11:36] <xnox> lovely
[11:38] <Laney> cjwatson: hmm no, I thought IS were going to subscribe everyone in that Launchpad team
[11:38] <Laney> I asked for the admins to be list mods, which includes you IIRC
[11:39] <Laney> (I'm subscribed; I got a welcome message, and vorlon replied to my test mail to the list)
[11:39] <paride> xnox, not really informative but this is what happens then the installed system tries to boot: https://people.ubuntu.com/~paride/misc/Screenshot.png
[11:39] <cjwatson> Laney: Sounds lovely but AFAICS I'm not
[11:40] <Laney> if you didn't see that mail, then I guess no indeed
[11:40] <cjwatson> e.g. hyperkitty just tells me "you must be subscribed to view the archives"
[11:40] <Laney> I guess I'll ping IS and ask them to double check, please stand by
[11:40] <cjwatson> ta
[11:41] <cjwatson> (huh, I'm apparently on ubuntu-archive-robot-owner@lists ...)
=== popey8 is now known as popey
[11:58] <Laney> there's a link I could click to confirm the change
[12:06] <xnox> paride:  ah, nah, that's fine, that's just broken lvm package.
[12:06] <xnox> paride:  will be fixed once https://launchpad.net/ubuntu/+source/lvm2/2.03.11-2ubuntu3 migrates.
[12:06] <Laney> we should revert lvm in release shouldn't we
[12:06] <xnox> seb128: slyon: looks like our installer tests did catch lvm2 bug as per paride above.
[12:07] <xnox> Laney:  probs. cause this lvm2 rebuild looks like stuck behind autopkgtests.
[12:07] <Laney> i'll do it
[12:07] <paride> xnox, does it make sense that bios installs work?
[12:07] <Laney> someone else can wrangle those tests
[12:08] <xnox> paride: with lvm?
[12:08]  * xnox is puzzled
[12:08] <xnox> yeah, some go thing fails to build.
[12:08] <slyon> xnox: that's good. But an lvm2 autopkgtest would probably be even better, to catch it earlier
[12:08] <paride> xnox, let me double check
[12:10] <Laney> it is done
[12:18] <xnox> Laney:  thank you
[12:20] <paride> xnox, bios install is also broken, not sure why it passes on the platform-qa jenkins
[12:20] <paride> but fails locally
[12:22] <xnox> paride:  good.
[13:01] <Laney> icey: hey could you look into cinder/armhf autopkgtest failure please? looks more likely to be container related than armhf related to me
[13:03] <icey> looking
[13:05] <Laney> think I'll add a badtest hint, so it'll need to be fixed for the next version
[13:06] <icey> Laney: in essence, it looks like it's from the new addition of python3-rtslib-fb, and that failing with a modprobe error
[13:06] <Laney> yup
[13:07] <Laney> if cinder is supposed to work inside a container, that is something to fix
[13:07] <Laney> otherwise Restrictions: isolation-machine is available to you :-)
[13:07] <icey> well, cinder doesn't require python3-rtslib-fb to work in all cases
[13:07] <Laney> or a ConditionVirtualization on the unit in question or something
[13:08] <icey> Laney: any idea how this was handled with tgt previously?
[13:08] <Laney> mostly for someone who knows the software to decide on :>
[13:08] <Laney> sorry afraid not
[13:08] <icey> ah no worries
[13:10] <icey> Laney: thanks for the poke, I have a thought on fixing it
[13:10] <Laney> great, thanks!
[13:11] <Laney> ok so lvm2 should go in next time as far as I can see
[13:24] <Laney> cjwatson: IS says the subscriptions should be fixed. I tried clicking the links myself but they don't work for me - guess you have to be logged in as the account being manipulated
[13:24]  * Laney vanishes for Mandatory Daily Walk
[13:27] <rbasak> sil2100: o/ looking at unattended-upgrades in the Xenial queue (carried over from yesterday)
[13:28] <cjwatson> Laney: Yeah, they're not
[13:28] <cjwatson> Something is weird
[13:31] -queuebot:#ubuntu-release- Unapproved: rejected unattended-upgrades [source] (xenial-proposed) [1.1ubuntu1.18.04.7~16.04.7]
[13:32] -queuebot:#ubuntu-release- Unapproved: accepted unattended-upgrades [source] (xenial-proposed) [1.1ubuntu1.18.04.7~16.04.7]
[13:43] <icey> hey Laney: do you know if there's a packaging thing that would allow installing a dependency if not in a container?
[14:23] <sil2100> rbasak: hey! o/ Ok! Though since it's Friday, it's Timo's SRU day o/ tjaalton ^
[14:23] <tjaalton> yup
[14:26] <rbasak> Oh, right. Sorry, I forgot what day it was.
[14:26] <rbasak> Anyway, it's done and accepted :)
[14:30] <tjaalton> that's fine :)
[14:47] <Laney> icey: nope, I think you need to make the thing itself deal with that using something like ConditionVirtualization if that's appropriate
[14:47] <icey> Laney: as it turns out, there's already a bug for this that I'm now working on: https://bugs.launchpad.net/ubuntu/+source/python-rtslib-fb/+bug/1865037:-D
[14:47] <ubot5> Ubuntu bug 1865037 in python-rtslib-fb (Ubuntu) "make the service fail gracefully if unable to load modules" [High,Triaged]
[14:50] <Laney> nothing gets past cpaelzer!
[14:58] -queuebot:#ubuntu-release- Unapproved: accepted libvirt [source] (focal-proposed) [6.0.0-0ubuntu8.8]
[15:00] <cjwatson> Laney: OK, fixed the mailman confusion and I've switched ubuntu-archive-robot's email address over now
[15:00] <cjwatson> So in theory there should start to be much email there
[15:02] <Laney> cjwatson: cheers. Looking forward to it.
[15:02] -queuebot:#ubuntu-release- Unapproved: accepted libvirt [source] (groovy-proposed) [6.6.0-1ubuntu3.4]
[15:05] -queuebot:#ubuntu-release- Unapproved: nvidia-graphics-drivers-390 (bionic-proposed/restricted) [390.141-0ubuntu0.18.04.1 => 390.141-0ubuntu0.18.04.2] (kernel-dkms, ubuntu-desktop)
[15:11] -queuebot:#ubuntu-release- Unapproved: accepted libvirt [source] (xenial-proposed) [1.3.1-1ubuntu10.31]
[15:21] -queuebot:#ubuntu-release- Unapproved: accepted opencryptoki [source] (groovy-proposed) [3.14.0+dfsg-0ubuntu3.1]
[15:22] -queuebot:#ubuntu-release- Unapproved: accepted opencryptoki [source] (focal-proposed) [3.13.0+dfsg-0ubuntu5.1]
[15:24] -queuebot:#ubuntu-release- Unapproved: accepted opencryptoki [source] (bionic-proposed) [3.9.0+dfsg-0ubuntu1.4]
[15:25] -queuebot:#ubuntu-release- Unapproved: accepted opencryptoki [source] (xenial-proposed) [3.4.1+dfsg-1ubuntu4.1]
[16:22] <Laney> ah there we go
[16:30] <cjwatson> indeed, spammity spam
[16:31] <cjwatson> err, shall I see if I can turn off the [ubuntu-archive-robot] prefix?
[16:32] <cjwatson> everyone is surely going to be filtering this by list-id or something anyway
[16:33] <cjwatson> done, will see if people complain
[16:33] <Laney> sure
[16:33] <Laney> I already heard from seb128 and didrocks that they unsubbed :-)
[16:34] <cjwatson> I'd suggest that people disable delivery rather than unsubbing - that way they still have access to look in the archives if they need to debug something
[16:35] <cjwatson> that being the point of this
=== _hc[m] is now known as _hc
[16:51] <RikMills> I would quite like to get the new flameshot (screenshot) tool into hirsute. Since that missed feature freeze by just 2 days, and is not seeded anywhere, is there a chance I could get that in without FFe paperwork?
[16:52] <RikMills> https://github.com/flameshot-org/flameshot/releases
[16:53] <RikMills> "Improved Wayland support on Gnome" is interesting
[17:40] -queuebot:#ubuntu-release- Unapproved: accepted initramfs-tools [source] (focal-proposed) [0.136ubuntu6.4]
[17:42] -queuebot:#ubuntu-release- Unapproved: accepted linux-firmware [source] (bionic-proposed) [1.173.20]
[17:50] -queuebot:#ubuntu-release- Unapproved: accepted nvidia-settings [source] (bionic-proposed) [460.39-0ubuntu0.18.04.2]
[17:53] -queuebot:#ubuntu-release- Unapproved: accepted nvidia-graphics-drivers-390 [source] (bionic-proposed) [390.141-0ubuntu0.18.04.2]
[18:53] -queuebot:#ubuntu-release- Unapproved: apt (groovy-proposed/main) [2.1.10ubuntu0.2 => 2.1.10ubuntu0.3] (core, i386-whitelist)
[19:02] <juliank> ^ launchpad generated the diff against 2.1.11 instead of 2.1.10ubuntu0.2, sorry for that
[19:02] <juliank> I don't know why I didn't go for 2.1.11ubuntu0.1 last update
[19:03] <juliank> that would fix the issue
[19:03] <juliank> 2.1.11 was a subset of the changes in 2.1.10ubuntu0.2
[19:03] <juliank> But it got overthrown by 2.1.10ubuntu0.1 security update, so I went on from there
[19:04] <juliank> now all the diffs are off
[19:04] <juliank> But at least it will be over in a few months :D
[19:04] <juliank> (suggest to use git-ubuntu for this SRU review)
[19:50] <vorlon> mwhudson: built on i386 is "ask an AA to build it and explain why it's wanted"
[19:59] <vorlon> mwhudson: is this needed for debcargo?
[21:02] <teward> cjwatson: vorlon: can you take a stab and check the iso builders to see if your syncs froze up again?  Lubuntu version mismatches after a forced rebuild
[21:02] <teward> unless the sync system hasn't run yet for the daily run
[21:02] <teward> cc wxl
[21:05] <cjwatson> Bleh
[21:05] <cjwatson> This is fundamentally about Laney's germinate timeout thing that I need to get round to finishing reviewing
[21:05] <cjwatson> In the meantime let me cancel some livefs builds and that should unstick things eventually
[21:07] <cjwatson> (livefs builds are getting stuck, so sets of image builds are never finishing, so the lock stays held)
[21:12] <wxl> sigh i guess i should have looked at the list. it's obvious it is still stuck on 1ubuntu40
[21:13] <wxl> i had hoped the issue was that the new grub had come almost right when we got our new iso so i was thinking maybe one package had migrated but not the other but i guess that's not the case
[21:13] <cjwatson> It's finishing up some slightly out-of-order image builds now but is getting there
[21:14] <cjwatson> I'll run new Lubuntu builds once the logjam has cleared
[21:14] <wxl> thanks cjwatson
[21:37] <cjwatson> merged that germinate fix (for bug 1912495) and uploaded to Debian; I'll sync it once it works its way through various pipelines.  somebody will still need to sort out SRUs, but that should at least start helping to lower the failure probability
[21:37] <ubot5> bug 1912495 in germinate (Ubuntu) "Can hang forever in downloading" [High,Fix committed] https://launchpad.net/bugs/1912495
[21:58] <cjwatson> teward,wxl: 20210305.2 is on its way out and looks better
[22:01] <wxl> thanks cjwatson
[22:02] -queuebot:#ubuntu-release- New binary: visp [s390x] (hirsute-proposed/universe) [3.4.0-1] (no packageset)
[22:07] <vorlon> apw_: lotsa kernel packages listed as reverse-build-deps of dh-systemd on https://people.canonical.com/~ubuntu-archive/nbs.html, can you update them to drop this build-dep? (since you already build-dep on debhelper-compat 10, and dh-systemd became a transitional package with debhelper 9.20160709, the B-D can just be dropped from bionic and later)
[22:08] -queuebot:#ubuntu-release- New binary: visp [ppc64el] (hirsute-proposed/universe) [3.4.0-1] (no packageset)
[22:08] <vorlon> apw_: (actually even the xenial version is newer)
[22:21] -queuebot:#ubuntu-release- New binary: visp [amd64] (hirsute-proposed/universe) [3.4.0-1] (no packageset)
[22:39] -queuebot:#ubuntu-release- New binary: visp [armhf] (hirsute-proposed/universe) [3.4.0-1] (no packageset)
[22:41] -queuebot:#ubuntu-release- New binary: visp [arm64] (hirsute-proposed/universe) [3.4.0-1] (no packageset)