| RAOF | ??? | 01:38 |
|---|---|---|
| RAOF | What is happening here? mir-1.7.0-0ubuntu2 has a libmiral3.symbols file which includes `(c++)mir::Foo::~Foo()`. objdump -T on libmiral.so.3 from that package asserts that it does *not* contain `mir::Foo:~Foo()`. | 01:40 |
| RAOF | And then the build of mir-1.7.1-0ubuntu1 with no changes to `Foo` fails because `mir::Foo::~Foo()` doesn't exist in the object? | 01:41 |
| wxl | did we add some thing to check the iso on boot? | 01:41 |
| wxl | if so, where? | 01:41 |
| RAOF | I don't see how all three of these things can happen?! | 01:41 |
| wxl | also does anyone know if this new iso manifest-diff check applies to flavors? or can we make it be? is there code for this somewhere? | 01:42 |
| wxl | er huh i should probably ask this on release blah | 01:43 |
| seb128 | vorlon, hey! so icu still didn't go as expected? :-( | 07:19 |
| vorlon | seb128: I think it's going to go in the current run | 07:19 |
| seb128 | fingers crossed! | 07:20 |
| vorlon | should've gone in a previous run but there was a znc sync and it took me a while to realize that was entangled with python3.8 | 07:20 |
| seb128 | hum, is there a problem with the arm64 autopkgtest infra/capacity? | 08:31 |
| seb128 | backlog doesn't seem to go down but we didn't upload that much in recent days? | 08:32 |
| seb128 | vorlon, I had packages just migrating but not those blocked on icu, I guess it means it failed again? :( | 08:38 |
| didrocks | it go down, but not fast. My yesterday’s upload only ran this morning on arm64 | 08:57 |
| seb128 | Laney, juliank, vorlon, do we have a page that list the number of active builders for autopkgtest? something like https://launchpad.net/builders/ | 09:07 |
| Laney | nope | 09:07 |
| seb128 | :( | 09:07 |
| Laney | you can count the builds on the running page | 09:07 |
| seb128 | the arm64 backlog doesn't seem to go down | 09:07 |
| seb128 | let me do that | 09:08 |
| juliank | seb128: patches welcome! | 09:08 |
| juliank | :D | 09:08 |
| seb128 | :) | 09:08 |
| Laney | ubuntu@juju-prod-ues-proposed-migration-machine-11:~$ journalctl --since=08:00 ADT_ARCH=arm64 | grep -c 'Acknowledging request' | 09:11 |
| Laney | 114 | 09:11 |
| Laney | it is going | 09:11 |
| LocutusOfBorg | rafaeldtinoco, ruby-psych and jruby have a chicken and egg serious problem, and now they are uninstallable both... | 10:11 |
| rafaeldtinoco | LocutusOfBorg: hum, that was part of the ruby transition uploads, I'm almost home will take a look shortly | 10:13 |
| LocutusOfBorg | thanks | 10:26 |
| rafaeldtinoco | kanashiro: fyi ^ | 10:40 |
| rafaeldtinoco | looks like for this build loop I'll have to drop ruby-psych from build-depends of jruby, and after building ruby-psych do another build of jruby supporting it | 11:02 |
| rafaeldtinoco | and open a bug in debian to check what they can come up with as ruby-psych has been alternating between suggest and dependency since 2017 | 11:03 |
| rafaeldtinoco | kanashiro: ^ | 11:03 |
| seb128 | icu migrated \o/ | 11:07 |
| seb128 | vorlon, well done :-) | 11:07 |
| kanashiro | rafaeldtinoco, let me know if your plan work as you planned | 11:14 |
| rafaeldtinoco | kanashiro: yep, will give it a try locally now | 11:14 |
| rafaeldtinoco | kanashiro: your transitions page look much better today | 11:19 |
| rafaeldtinoco | just a few packages missing from the 1st wave | 11:19 |
| kanashiro | rafaeldtinoco, yes, and I am checking it and some of the packages in red already built fine, it might be a matter of waiting to update it | 11:20 |
| rbasak | ahasenack: pam_mysql is a sync I believe. Could you please review before I sync it? | 11:29 |
| rbasak | (pam-mysql) | 11:29 |
| rbasak | libdbi-drivers is in main. So why doesn't https://people.canonical.com/~ubuntu-archive/germinate-output/ubuntu.focal/rdepends/libdbi-drivers/ exist? | 11:45 |
| seb128 | vorlon, your tepl upload to pick i386 is depwaiting on amtk to be also added | 12:41 |
| Laney | wish I could update that i386 seed myself | 12:43 |
| === wgrant is now known as wgrant_ | ||
| === wgrant_ is now known as wgrant | ||
| rbasak | Ah - apparently we unseeded libdbi-drivers this cycle | 13:40 |
| ahasenack | how can I get access to focal on armhf to troubleshoot http://autopkgtest.ubuntu.com/packages/c/cacti/focal/armhf ? | 15:04 |
| ahasenack | should I use a bileto ticket and do "interactive" debugging with it, one print() per upload? | 15:04 |
| Laney | ahasenack: get an arm64 VM and use lxd to launch armhf inside that | 15:16 |
| ddstreet | juliank hi, i'm assuming you probably won't have time to review this MR until after focal? https://code.launchpad.net/~ubuntu-support-team/software-properties/+git/software-properties/+merge/379928 | 15:54 |
| ddstreet | i can split it up or adjust or whatever, but need to know what you'd like to see done, otherwise i'll just keep developing and it will grow larger :) | 15:55 |
| juliank | ddstreet: that's hard | 16:00 |
| ddstreet | you mean it's hard to review at all? | 16:00 |
| juliank | i don't know | 16:01 |
| ddstreet | oh | 16:01 |
| ddstreet | well i'll be at the sprint next week if you want to talk | 16:01 |
| juliank | I think I reviewed in the wrong order | 16:01 |
| juliank | I was looking at the commit that removes PPA stuff before the commit that rewrites it | 16:02 |
| juliank | ddstreet: Do you know if anything is using the library that could break by this change? | 16:02 |
| cjwatson | ddstreet: https://code.launchpad.net/~ddstreet/launchpadlib/nosudo/+merge/379694 is on my queue but I'd find it helpful if you didn't add extra review requests please :) | 16:03 |
| ddstreet | cjwatson ack, i wasn't sure if you were on the list | 16:04 |
| juliank | ddstreet: I don't want it to get even larger :D | 16:04 |
| * cjwatson deletes that request - I'm already in ~lazr-developers | 16:04 | |
| juliank | but I also don't really have time or nerve to review this yet | 16:04 |
| ddstreet | juliank i dont think anything is using the changed code, but i'll check all the reverse-depends | 16:04 |
| juliank | worst case, I guess we can sit down for 30 mins next week and go over the commits | 16:05 |
| ddstreet | sounds good, thnx | 16:06 |
| === jdstrand_ is now known as jdstrand | ||
| ahasenack | Laney: thanks, that worked! | 16:23 |
| * Laney does a celebratory moonwalk | 16:24 | |
| === alan_g is now known as alan_g_ | ||
| === alan_g_ is now known as alan_g | ||
| vorlon | seb128: thanks for the hint re: amtk, sorted that now also | 20:15 |
| seb128 | vorlon, thanks for sorting it out, days are a bit crazy with ff this week but I will take time to learn to do that myself next time I get one | 20:15 |
| === hggdh is now known as hggdh-msft | ||
| seb128 | vorlon, what issue was your libxml2 recent patch fixing? There is no bug reference in the changelog/patch and Debian has applied what look like a subset of your change in https://packages.qa.debian.org/libx/libxml2/news/20200227T184953Z.html and I would like to check if that's enough to fix the issue you were having | 22:33 |
| seb128 | vorlon, in which case we could sync back, which would be nice because Debian also backported some extra CVE fixes | 22:34 |
| vorlon | seb128: I actually opened a PR upstream for my change, and upstream pointed out it was fixed differently on trunk. so a newer upstream is probably fine | 22:34 |
| vorlon | the issue was a testsuite failure in libxml-libxslt-perl on armhf | 22:35 |
| vorlon | because it was constructing trees of nodes that were unexpected by libxml2 | 22:35 |
| seb128 | vorlon, ok, Debian took the fix from upstream so I guess it should be fine to sync, thanks | 22:35 |
| seb128 | vorlon, right, https://salsa.debian.org/xml-sgml-team/libxml2/-/commit/f91d1425 sounds like a variant for the same fix | 22:36 |
| seb128 | thanks for confirming! | 22:36 |
| vorlon | and I spoke w/ mapreri this morning on #debian-devel | 22:36 |
| vorlon | so yeah we should be good | 22:36 |
| vorlon | if it gets hung up again on the autopkgtest, we'll know soon enough | 22:37 |
| seb128 | right | 22:37 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!