[00:03] -queuebot:#ubuntu-release- Unapproved: accepted ibus-braille [source] (kinetic-proposed) [0.3-7ubuntu22.10.1]
[00:04] -queuebot:#ubuntu-release- Unapproved: accepted ibus-braille [source] (jammy-proposed) [0.3-7ubuntu22.04.1]
[00:06] -queuebot:#ubuntu-release- Unapproved: accepted pulseaudio [source] (kinetic-proposed) [1:16.1+dfsg1-1ubuntu3.1]
[00:09] -queuebot:#ubuntu-release- Unapproved: accepted pulseaudio [source] (jammy-proposed) [1:15.99.1+dfsg1-1ubuntu2.1]
[00:54] -queuebot:#ubuntu-release- Unapproved: accepted openldap [source] (kinetic-proposed) [2.5.14+dfsg-0ubuntu0.22.10.1]
[01:11] -queuebot:#ubuntu-release- Unapproved: accepted openldap [source] (jammy-proposed) [2.5.14+dfsg-0ubuntu0.22.04.1]
[01:16] <sergiodj> hm, is something going on with the ppc64el testers?  I'm getting "linux-generic : Depends: linux-image-generic (= 6.1.0.16.16) but 5.19.0.21.21 is to be installed" when I retrigger tests for flycheck
[01:17] <sergiodj> https://autopkgtest.ubuntu.com/packages/flycheck/lunar/ppc64el
[01:43] <vorlon> let's see
[01:43] -queuebot:#ubuntu-release- Unapproved: accepted cloud-init [source] (kinetic-proposed) [23.1.1-0ubuntu0~22.10.1]
[01:44] <vorlon> sergiodj: rmadison -s lunar -S linux-meta -a ppc64el shows correct output
[01:44] <sergiodj> vorlon: yes, I noticed that too.  I couldn't figure out why I'm getting the error
[01:44] <sergiodj> there seems to be some stale cache somewhere...?
[01:45] <vorlon> maybe something has the old version held
[01:45] <sergiodj> I'm not actively looking into it right now, but I can do it later if it isn't solved by then
[01:46] <vorlon> I'm taking a brief look, but if there's nothing obvious I'm going to have to punt to paride on monday
[01:47] <sergiodj> sounds good.  thanks
[01:48] <vorlon> sergiodj: autopkgtest-cloud git shows no relevant changes in this timeframe and https://autopkgtest.ubuntu.com/results/autopkgtest-lunar/lunar/ppc64el/f/flycheck/20230303_050502_8a444@/log.gz shows the correct packages being pulled in not too long ago
[01:49] <sergiodj> vorlon: the plot thickens
[01:51] <sergiodj> there is something strange at play here.  I can confirm that it was possible to trigger flycheck tests on ppc64el earlier today
[01:51] <sergiodj> I will see if I can get a canonistack VM to investigate
[01:51] <vorlon> nothing obvious cowboyed onto the infra
[01:53] <vorlon> yeah I'm punting, sorry
[01:53] <sergiodj> this is the only test remaining for emacs to migrate.  sigh
[01:53] <sergiodj> vorlon: np, thanks for taking a look
[01:54] -queuebot:#ubuntu-release- Unapproved: accepted cloud-init [source] (jammy-proposed) [23.1.1-0ubuntu0~22.04.1]
[01:55] -queuebot:#ubuntu-release- Unapproved: accepted cloud-init [source] (focal-proposed) [23.1.1-0ubuntu0~20.04.1]
[01:58] -queuebot:#ubuntu-release- Unapproved: accepted cloud-init [source] (bionic-proposed) [23.1.1-0ubuntu0~18.04.1]
=== chris14_ is now known as chris14
[04:41] -queuebot:#ubuntu-release- Unapproved: apache2 (jammy-proposed/main) [2.4.52-1ubuntu4.3 => 2.4.52-1ubuntu4.4] (i386-whitelist, ubuntu-server)
[04:42] -queuebot:#ubuntu-release- Unapproved: apache2 (kinetic-proposed/main) [2.4.54-2ubuntu1.1 => 2.4.54-2ubuntu1.2] (i386-whitelist, ubuntu-server)
[07:08] <vorlon> juliank: why is LP: #1995751 listed as in-progress for bionic? (noticed because launchpad gave me a diff against 2.02-2ubuntu8.23 instead of 2.02-2ubuntu8.25 and the bug is in the changelog diff)
[07:08] -ubottu:#ubuntu-release- Launchpad bug 1995751 in grub2-unsigned (Ubuntu Bionic) "update to 2.04-1ubuntu47.4 drops zz-update-grub" [Undecided, In Progress] https://launchpad.net/bugs/1995751
[07:14] <vorlon> all image builds currently failing with an error connecting to ftpmaster.internal.  I have pinged Canonical IS about it but am not currently escalating it as an emergency
[07:16] -queuebot:#ubuntu-release- Unapproved: accepted grub2 [source] (bionic-proposed) [2.02-2ubuntu8.26]
[07:16] <juliank> vorlon: there's a dependency bump in grub2-unsigned 2.04 upload that it didn't get because it moved to 2.06 which I guess has the same bump but no changelog entry (would need double checking)
[07:17] <vorlon> ah
[07:17] <vorlon> well, accepted
[07:17] <vorlon> I had assumed the debian/control changes were all in .25 without actually checking :/
[07:17] <juliank> vorlon: the grub2 is fine, the unsigned could be buggy
[07:18] <juliank> As in if you upgrade with an older version of grub2 (< 25k installed
[07:19] <juliank> That was just a safety feature to make sure the grub2 update is pulled in if you update grub2-signed
[07:19] <vorlon> right
[07:20] <juliank> Thanks for accepting
=== guiverc2 is now known as guivec
=== guivec is now known as guiverc
[08:01] <vorlon> something non-obvious going on with boost1.74 migration, pycuda is an implicit dependency but not being included by the hinter.  I'm going to add a 'hint boost1.74' to see if it manages to get there
[08:27] <jamespage> vorlon: those are all good quesions - I don't have the full history but tl;dr some of our openstack charm code relies on this and it was non-functional unless you used a PPA - with a broken version in the archive
[08:27] <jamespage> I put aside some time at the request of the MAAS team to update and resolve some of those challenges - I will look at the autopkgtests in devel as well next week
[08:55] <vorlon> jamespage: cheers
[08:56] <vorlon> sergiodj: https://people.canonical.com/~ubuntu-archive/proposed-migration/lunar_uninst.txt shows linux-image-generic uninstallable on ppc64el and s390x but I can't reproduce this in a chroot
[09:04] <vorlon> sergiodj: https://launchpad.net/ubuntu/lunar/ppc64el/linux-image-6.1.0-16-generic suggests it was missing until I attempted a copy-back around 2 hours ago, however
[09:05] <vorlon> sergiodj: so I'm guessing the report above is just out-of-date, and the reason it was missing was from my accidental NBS removal, followed by me only doing a copy-back before of linux-signed, not linux
[17:54] <vorlon> well, the pycuda failure is still not making any sense in the output, it looks fine in update_excuses and a naive apt install works. I'm going to remove it from release pocket temporarily, which should have the side effect of giving us better explanations on the next run
[17:59] <vorlon> jbicha: glib and webkitgtk are both blocking NBS removal of icu.  What's the path forward for resolving these ongoing transitions in -proposed?  Should we roll back to let other packages through?
[18:02] <vorlon> giving back a batch of ppc64el test failures caused by the linux-image mess
[18:10] <vorlon> cyrus-imapd has a reproducible build failure on armhf
[18:21] <vorlon> bluesabre: is xubuntu-minimal now where you expect it to be size-wise? I see the 22.10 xubuntu core download was 1.3G and it looks like the xubuntu-minimal daily is 1.7G
[18:26] <bluesabre> vorlon: Not quite yet. I’m working to identify what else needs to be removed or adjusted. Thanks for checking in!
[18:41] <vorlon> bluesabre: ok, let me know if there's anything you need from my side
[18:42] <vorlon> I'm always in favor of reducing the size of images ;)  things are actually getting quite tight now on the infrastructure side wrt the size of the image set
[18:42] <vorlon> (we will be able to clean up some space once we get 20.04.6 done, fortunately, since we can then turn off focal dailies)
[19:24] <vorlon> whatever is going on with ftpmaster.internal is also intermittently affecting package builds
[20:45] <sergiodj> vorlon: got it.  thanks for investigating and fixing the issue
[20:51] <cjwatson> vorlon: apt-mirror/dists> coincidentally as of a couple of days ago LP has started retaining enough information to be able to reconstruct most of that from here on (well, only for Packages/Sources/InRelease and similar as yet, not for custom uploads)
[21:55] <vorlon> xnox: congrats, all but 1 of the oldest 14 packages in lunar-proposed are now kernel-related
=== chris15 is now known as chris14
[23:30] <jbicha> vorlon: you can temporarily drop webkit2gtk & glib2.0; we will need new versions of both of those anyway
[23:31] <jbicha> vorlon: please use webkit2gtk 2.38.5-1 since it's already been built against the new icu; and use gnome-builder 43.6-1
[23:32] <jbicha> we haven't isolated the riscv64 fix for webkit 2.39 yet; glib 2.75 has some regressions
=== chris14_ is now known as chris14
[23:45] <vorlon> jbicha: ack. I'll work on rolling those back later today