| jbicha | mwhudson: it's because network-manager isn't ready to migrate and it's part of the ppp transition | 00:20 |
|---|---|---|
| -queuebot:#ubuntu-release- Unapproved: accepted php7.2 [source] (bionic-proposed) [7.2.24-0ubuntu0.18.04.9] | 00:45 | |
| mwhudson | jbicha, bdmurray: ah ha | 01:22 |
| mwhudson | oh look netplan and systemd | 01:23 |
| -queuebot:#ubuntu-release- Unapproved: accepted systemd [source] (bionic-proposed) [237-3ubuntu10.52] | 01:35 | |
| mwhudson | bdmurray, jbicha: is the netplan failure just network-manager changing its output or something deeper i wonder | 02:14 |
| mwhudson | hmm probably not that simple | 02:39 |
| mwhudson | grar ok i think i have fixes for both of these failures | 04:40 |
| mwhudson | nothing like three uploads of systemd in three days | 04:47 |
| -queuebot:#ubuntu-release- Unapproved: accepted systemd [source] (hirsute-proposed) [247.3-3ubuntu3.5] | 05:42 | |
| mwhudson | ok netplan and systemd uploaded that will hopefully work with each other and the network-manager in proposed | 07:24 |
| === didrocks999 is now known as didrocks | ||
| laney | wasn't there some breakage between the new glibc and systemd that needed fixing in systemd? | 08:14 |
| laney | ah I guess it didn't migrate yet so it's ok | 08:15 |
| mwhudson | laney: which breakage do you mean? i've fixed at least some things like that... | 08:25 |
| laney | mwhudson: I was thinking of the change in ubuntu5/6 and how glibc was trying to migrate without systemd | 08:26 |
| mwhudson | ah | 08:26 |
| mwhudson | https://people.canonical.com/~ubuntu-archive/proposed-migration/update_output_notest.txt shows glibc trying to migrate with systemd | 08:27 |
| mwhudson | and failing because of that rando ruby package | 08:27 |
| laney | extra test triggers should make britney insert its own dependency that ties things together, then it would happen at the excuses stage | 08:29 |
| laney | yummy | 08:29 |
| laney | ruby-mizuho ? | 08:31 |
| schopin | for my education, could you tell me where to look in this file to come out with the conclusion regarding glibc being blocked by a ruby package? | 08:31 |
| laney | schopin: it's a bit of art, but in this case search for 'glibc/' and look below the huge list of packages | 08:32 |
| schopin | Oh, OK. So basically britney tries to migrate everything in one big batch, this fails because of ruby-mizuho, and because of that it tries the packages one by one. | 08:33 |
| laney | I think it tries them one by one first ("trying: glibc" is earlier in the file) | 08:33 |
| laney | but yeah, then it's got some smarts to try to calculate what should go together | 08:33 |
| schopin | First mention I get is in the huge list :/ | 08:34 |
| schopin | Thanks for the explanation! | 08:35 |
| laney | mwhudson: do you have a handle on that mizuho thing? | 08:37 |
| mwhudson | laney: not really | 08:40 |
| mwhudson | it's because the new libxml2 declares a breaks: on something | 08:40 |
| laney | yeah, ruby-nokogiri, we should take that I guess? | 08:50 |
| -queuebot:#ubuntu-release- Unapproved: rejected curl [source] (focal-proposed) [7.68.0-1ubuntu2.7] | 08:55 | |
| mwhudson | laney: yeah there's a new one in experimental, hopefully it doesn't break any of the 1 million reverse build depends | 09:01 |
| mwhudson | (it's some kind of documentation generator?) | 09:01 |
| laney | something like that | 09:04 |
| laney | mapreri: any insight as the person who added the breaks? | 09:04 |
| seb128 | mwhudson, https://github.com/sparklemotion/nokogiri/issues/2257 seems the reason, maybe just backport that fix if you don't want to take the experimental version? | 09:08 |
| ubottu | Issue 2257 in sparklemotion/nokogiri "[bug] 1.11.4 - New libxml2 recursion limit breaks parsing for complex files" [Closed] | 09:08 |
| seb128 | the patch is simple https://github.com/sparklemotion/nokogiri/commit/7c10c01477 | 09:09 |
| ubottu | Commit 7c10c01 in sparklemotion/nokogiri "fix: upstream libxml2 bug in calculating xpath query recursion depth" | 09:09 |
| seb128 | we would need an libxml upload to update the breaks version | 09:10 |
| mwhudson | i guess looking at the diff between what we have and what is in experimental would be a good idea | 09:15 |
| mwhudson | uh, it is large | 09:18 |
| mwhudson | wait, that fix is a patch to libxml2 itself? | 09:19 |
| laney | yes | 09:21 |
| laney | embedded copy or something | 09:21 |
| laney | there's a link to the upstream fix in the issue | 09:21 |
| mwhudson | ok so we can do a libxml2 upload adding that patch and dropping the breaks | 09:26 |
| mwhudson | it applies cleanly enough... | 09:33 |
| mwhudson | laney, seb128, mapreri: look ok? https://paste.ubuntu.com/p/wx2KdSxssS/ | 09:33 |
| laney | lgtm but I don't know about the specific problem being broken/fixed by this | 09:34 |
| mwhudson | yeah, i'm just trusting upstreams basically | 09:35 |
| mwhudson | maybe i'll sleep on it :) | 09:35 |
| laney | yolo iy | 09:38 |
| laney | it* | 09:38 |
| ginggs | mwhudson: mapreri asks if you can wait | 09:41 |
| ginggs | he'll pick that in debian and upload to unstable later today | 09:42 |
| mwhudson | ginggs: ok | 09:42 |
| ginggs | mwhudson: sleep well! | 09:44 |
| laney | where's the side conversation happening? :O | 09:45 |
| ginggs | laney: mapreri is afk, i have him on telegram | 09:46 |
| laney | ah | 09:46 |
| mwhudson | if someone could look at https://bugs.launchpad.net/ubuntu/+source/docker.io/+bug/1942276 and try to understand what is going on, that would be gret | 09:54 |
| ubottu | Launchpad bug 1942276 in docker.io (Ubuntu) "docker 20.10.7-0ubuntu2 fails to start when installed inside Impish LXD container" [Medium, New] | 09:54 |
| mwhudson | *great | 09:54 |
| mwhudson | (and this is possibly a good reason to not let glibc migrate just yet) | 09:54 |
| ginggs | systemd + network-mananger passed together https://autopkgtest.ubuntu.com/packages/s/systemd/impish/arm64 | 10:21 |
| ginggs | but cryptsetup seems bad | 10:21 |
| -queuebot:#ubuntu-release- Unapproved: llvm-toolchain-12 (focal-proposed/main) [1:12.0.0-3ubuntu1~20.04.3 => 1:12.0.0-3ubuntu1~20.04.4] (i386-whitelist) | 10:35 | |
| -queuebot:#ubuntu-release- Unapproved: llvm-toolchain-12 (hirsute-proposed/universe) [1:12.0.0-3ubuntu1~21.04.1 => 1:12.0.0-3ubuntu1~21.04.2] (i386-whitelist) | 10:38 | |
| xnox | apw: please process binary-only removals of raspi ABI that got superseded in impish-proposed as listed under https://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html#linux-raspi | 10:41 |
| xnox | old binaries left on arm64: linux-buildinfo-5.13.0-1005-raspi, linux-headers-5.13.0-1005-raspi, linux-image-5.13.0-1005-raspi, linux-modules-5.13.0-1005-raspi, linux-modules-extra-5.13.0-1005-raspi, linux-raspi-headers-5.13.0-1005, linux-raspi-tools-5.13.0-1005, linux-tools-5.13.0-1005-raspi (from 5.13.0-1005.6) | 10:41 |
| xnox | old binaries left on armhf: linux-buildinfo-5.13.0-1005-raspi, linux-buildinfo-5.13.0-1005-raspi-nolpae, linux-headers-5.13.0-1005-raspi, linux-headers-5.13.0-1005-raspi-nolpae, linux-image-5.13.0-1005-raspi, linux-image-5.13.0-1005-raspi-nolpae, linux-modules-5.13.0-1005-raspi, linux-modules-5.13.0-1005-raspi-nolpae, linux-modules-extra-5.13.0-1005-raspi, | 10:41 |
| xnox | linux-modules-extra-5.13.0-1005-raspi-nolpae, linux-raspi-headers-5.13.0-1005, linux-raspi-tools-5.13.0-1005, linux-tools-5.13.0-1005-raspi, linux-tools-5.13.0-1005-raspi-nolpae (from 5.13.0-1005.6) | 10:41 |
| xnox | then with arm64 override, raspi should be able to migrate | 10:41 |
| LocutusOfBorg | tjaalton, ^^ please? llvm-toolchain-12 | 10:45 |
| cpaelzer | arr - these words are too overloaded for search engines, and my memory too bad. Was there another way than adding a no-op -Xubuntu1 upload to prevent syncing of a package once we open for 22.04? | 11:25 |
| rbasak | There is a "never autosync" list but I think entries in there are supposed to be permanent | 11:27 |
| cpaelzer | well, I want to resolve it some day, but plenty of rust things prevent me short term | 11:28 |
| rbasak | https://bazaar.launchpad.net/~ubuntu-archive/+junk/sync-blacklist/view/head:/sync-blacklist.txt | 11:29 |
| cpaelzer | I guess it is auto-sync from ubuntu-archive-tools so I can take a look there | 11:29 |
| cpaelzer | ah you found it while I'm typing, thanks rbasak | 11:29 |
| cpaelzer | :-) while checking out I realized I had it on disk already :-) | 11:32 |
| cjwatson | The sync-blacklist isn't necessarily permanent IME | 11:33 |
| cpaelzer | thanks, I'll consider filing an MP thre once I'm sure | 11:38 |
| cjwatson | You can't file an MP since it's a +junk branch | 11:38 |
| cjwatson | (That likely needs to be changed at some point, preferably along with moving it to git) | 11:39 |
| cpaelzer | oh indeed, thanks for the hint | 11:40 |
| cpaelzer | one less time bzr will drive me mad via +junk | 11:40 |
| apw | xnox, NBS wackage | 13:12 |
| bdmurray | ginggs: IIUC we need a new version of libxml2 to help with the glibc migration, is that right? | 16:02 |
| ginggs | bdmurray: that's correct | 16:03 |
| ginggs | 2.9.12+dfsg-3 already uploaded to unstable | 16:04 |
| ginggs | can be sync'd as soon as launchpad knows about it | 16:04 |
| bdmurray | ginggs: okay, let me know if you don't get to it and I can keep an eye out | 16:06 |
| ginggs | bdmurray: ok, thanks | 16:07 |
| ginggs | bdmurray: it doesn't look like libxml2 was picked up in the last import, so I guess it will be in ~6 hours. will you sync please? | 18:34 |
| bdmurray | ginggs: will do | 18:35 |
| ginggs | ta! | 18:35 |
| kanashiro | could anyone from the release team acknowledge this FFe request? https://bugs.launchpad.net/ubuntu/+source/fence-agents/+bug/1942368 | 19:55 |
| ubottu | Launchpad bug 1942368 in fence-agents (Ubuntu) "[FFe] Move fence_mpath to the -base package" [Undecided, New] | 19:55 |
| kanashiro | it is a very simple change | 19:55 |
| bdmurray | done | 20:12 |
| kanashiro | bdmurray, thanks! | 20:24 |
| bdmurray | mwhudson: I've sync'ed libxml | 22:36 |
| mwhudson | bdmurray: thanks | 22:37 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!