[00:21] PR snapcraft#3233 opened: travis: set cla check as final stage [03:23] https://forum.snapcraft.io/t/get-snap-package-and-dependencies-size-before-installing/19135 [03:23] https://forum.snapcraft.io/t/how-to-get-dependencies-and-reserve-dependencies-for-a-sanp-package/19134 [05:14] morning [07:17] mborzecki: hi, seems we got a new strange centos failure? https://github.com/snapcore/snapd/pull/9068 [07:17] PR #9068: tests/main/selinux-clean: workaround SELinux denials triggered by linger setup on Centos8 [07:19] pedronis: hi, hmmm that's weird [07:19] it's not related to your change but yes [07:22] huh, /sys/fs/cgroup/unified really isn't there [07:22] PR snapd#9006 closed: bootloader: compose command line with mode and extra arguments [07:22] PR snapd#9068 closed: tests/main/selinux-clean: workaround SELinux denials triggered by linger setup on Centos8 === pedronis_ is now known as pedronis [07:24] pedronis: systemd was switched to use default-hierarchy=legacy [07:25] interesting [07:27] PR snapd#9071 closed: release: 2.45.3.1 [07:27] PR snapd#9072 closed: packaging: add placeholder changelog for 2.45.3.1 [07:28] mborzecki: I merged the various release related branches, and I uploaded the release source package [07:28] to the ppa [07:29] pedronis: great, thank you for working on it [07:29] let's see what the build will contain [07:29] as for centos8, something is off, there is a changelog entry for systemd, but the switch to legacy was back in 2018 in version 239-7 [07:30] mborzecki: the changelog here now looks right: https://launchpad.net/~snappy-dev/+archive/ubuntu/image/+packages [07:30] let's say how the build goes [07:30] we still have the named systemd hierarchy [07:31] s/say/see [07:32] pedronis: hmm, the diff is large [07:33] ah right, it's a diff between 2.45.3 upload (whic was really master) and 2.45.3.1 [07:34] and it removes things that were on master, so looks legit [07:37] mborzecki: are you blocked on me on something atm? [07:37] pedronis: no, i'm good [07:37] quick errand, delivery [07:38] or not [07:40] mborzecki: when you have a moment, could you review #9067 ? (UC20 related) [07:40] PR #9067: cmd/snap-bootstrap/initramfs-mounts: add doSystemdMount + unit tests [07:41] pedronis: yeah, have it open since yesterday :/ [07:46] mborzecki: so amd64 build worked, but we have some build failures (test timeouts) [07:49] retry? [07:49] yes, that's what I'm doing [08:15] there's quite a bit of conflicts merging back into master :/ [08:15] pedronis: #9063 is not waiting for anything, we can land it right? [08:15] PR #9063: cmd/snap/debug: add "snap debug seeding" command for preseeding debugging (4/N) [08:15] mborzecki: I think so [08:16] mborzecki: btw there were few changes on master to arch PKGBUILD hope they aren't mandatory [08:17] pedronis: i can take a look when you open the 2.45->master merge branch [08:18] PR snapd#9063 closed: cmd/snap/debug: add "snap debug seeding" command for preseeding debugging (4/N) [08:19] mborzecki: there's 20 files in conflict, might take a while [08:20] please could somebody have a look at the snapd autopkg test regressions triggered by dpkg? [08:36] doko: do you have a link to the test? [08:38] mborzecki: https://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html [08:40] hmm that should be an easy fix hopefully [08:42] quick errand, this time for real [09:23] PR snapd#9073 opened: release: 2.45.3.1 [09:41] re [09:46] doko: this should be fixed in 2.45.3 [09:47] and .3.1 too [10:33] PR snapd#9074 opened: bootloader: extend managed assets bootloader interface to compose a candidate command line [11:51] pedronis: do you know how the vendor, no_vendor tarballs get generated for the github releases page? [11:53] mborzecki: not really, but I would imagine no vendor can be created with git archive maybe? [11:55] pedronis: hm my guess it's from slicing the tarball we have in LP, mvo usually uploaded those after the builds were done [11:55] mborzecki: do we have a tarball in LP? [11:56] pedronis: there's the source tarball that was uploaded for build [11:56] there is, that has vendor in though [11:57] ha, w8 i think i got something [12:04] morning folks [12:05] seems that #9063 was landed without some of the changes that I made in #9035 [12:05] PR #9063: cmd/snap/debug: add "snap debug seeding" command for preseeding debugging (4/N) [12:05] PR #9035: o/devicestate: save seeding/preseeding times for use with debug seeding api (3/N) [12:05] shall I merge master into #9035 or just open a separate PR to make for simpler git history ? [12:05] 9035 is also green, so perhaps that would be quickest to just get another +1 to that PR [12:17] ijohnson: hey, i'll take a look [12:18] thanks mborzecki [12:18] ijohnson: maybe you could do a pass over #9074 [12:18] PR #9074: bootloader: extend managed assets bootloader interface to compose a candidate command line [12:18] it's just two patches after pedronis' review fixing error message and tests [12:18] mborzecki: sure I will add it to my queue [12:18] ijohnson: oh, and i fixed the caching & artifact [12:18] mborzecki: yes I saw that, thanks for keeping up with that [12:18] seems my idea last night was a bit too naive [12:19] fwiw 33MB of a snap is ok to fit in the cache ;) [12:20] wonder how large the files can be and if we can use that in an interesting ways maybe [12:23] mborzecki: did you find out how to make those tars? [12:24] pedronis: sorry, yes, there's a release-tools/repack-debian-tarball.sh script in snapd source tree which does the slicing and stitching [12:26] pedronis: ./repack-debian-tarball.sh snapd_2.45.3.1.tar.xz will produce 3 tarballs [12:27] ah, good [12:29] PR snapd#9035 closed: o/devicestate: save seeding/preseeding times for use with debug seeding api (3/N) [12:29] PR snapd#9075 opened: daemon/api: use pointers to time.Time for debug seeding aspect [12:29] thanks mborzecki I just merged it [12:29] and opened the final bit for snap debug seeding, 9075, which is very simple [12:33] ijohnson: https://github.com/snapcore/snapd/pull/9067#discussion_r462261549 little tweak [12:33] PR #9067: cmd/snap-bootstrap/initramfs-mounts: add doSystemdMount + unit tests [12:35] mborzecki: thanks! I will take that into a followup I think since this is green [12:35] * ijohnson really likes how green things are to master as of late [12:35] * ijohnson (with much kicking of github actions) [12:39] PR snapd#9067 closed: cmd/snap-bootstrap/initramfs-mounts: add doSystemdMount + unit tests [13:09] PR snapd#8558 closed: tests: make the nested library usable independently of spread [13:48] * cachio afk -> buy a medicine [13:50] pedronis: #8867 is sudo git mergable, the tests that failed are unrelated, but we could also try to make it green with a master merge [13:50] PR #8867: interfaces: add uinput interface <⛔ Blocked> [14:11] ijohnson: did you look at it, it changed a bit since it was last reviewed by others but me? [14:12] pedronis: sorry no I didn't, do you want me to do another review since it changed ? [14:12] ijohnson: yes, also maybe a master merge is in order, though I think is quite self contained [14:12] pedronis: ok, I will do a master merge and review it then [14:13] ijohnson: thank you [14:25] pedronis: I approved that PR and merged master there [14:27] cmatsuoka: hmm i forgot we have ModeRun/ModeRecover in the boot pacakge [14:27] PR snapcraft#3233 closed: travis: set cla check as final stage [14:32] 8-line test only PR that only needs 1 more +1: https://github.com/snapcore/snapd/pull/8998 [14:32] PR #8998: tests/cmd/snap-bootstrap/initramfs-mounts: add test case for empty recovery_mode [14:33] also pedronis do you need me to review 9073 too ? [14:33] mborzecki: :) [14:34] ijohnson: yes if you can [14:34] pedronis: ack will do [14:35] cmatsuoka: btw can you try your branch with the code from #9074 ? [14:35] PR #9074: bootloader: extend managed assets bootloader interface to compose a candidate command line [16:29] do we still have the limitation that content interfaces can only be consumed if both snaps come from the same publishr (by default that is ... ) ? [16:29] ogra_: that limitation is only about _auto-connection_ [16:30] ogra_: you can have a content interface from snap from publisher A and another snap from publisher B use that content snap, but either you need to get an auto-connection from the store or make the connection manually [16:31] okay, thanks ... i guess that means a store-request then [16:33] PR snapcraft#3231 closed: file utils: rename get_tool_path() to get_snap_tool_path() [16:53] PR snapcraft#3234 opened: project loader: refactor errors to conform to SnapcraftException [17:03] PR snapcraft#3235 opened: tests: fix assert ordering for error format tests [17:15] PR snapd#8867 closed: interfaces: add uinput interface === ijohnson is now known as ijohnson|lunch [18:38] PR snapcraft#3225 closed: specifications: default tracks [18:45] * cachio kinesiologist [20:18] PR snapcraft#3236 opened: snap: use python3-apt stage-package [22:11] PR snapd#8932 closed: o/ifacestate: update security profiles in connect undo handler