=== Ringtailed_Fox is now known as RingtailedFox [00:53] PR snapcraft#3226 closed: extensions: kde-neon: use gtk3 platform theme on gtk-based DE's [02:05] PR snapd#9067 opened: cmd/snap-bootstrap/initramfs-mounts: add doSystemdMount + unit tests === benfrancis5 is now known as benfrancis [04:46] morning [06:23] hello [06:23] I tried to dput the .3 release again at 4AM but it was rejected, - the orig tarball is not removed from te ppa yet [06:24] I wonder if we could ask someone from launchpad for help [06:24] I can upload the signed release somewhere so that after the turmoil ends someone else can dput it [06:24] mborzecki: ^ [06:25] zyga: hey, what do i need to dput it? [06:25] I've sent a copy to maciek [06:25] you need dput [06:25] the magic line is [06:26] dput ppa:snappy-dev/image snapd_2.45.3_source.changes [06:26] brb, let me get a shower [06:26] I'm leaving at 11:30 [06:35] I can try to dput now [06:35] and maybe you can ask in #launchpad [06:36] if they can somehow speed up the removal / garbage collection of the 2.45.3 release leftovers [06:36] I will dput now [06:36] dput always works if the release is valid [06:36] but an email sent later [06:36] PR snapd#9068 opened: tests/main/selinux-clean: workaround SELinux denials triggered by linger setup on Centos8 [06:36] tells you if you were successful or not [06:36] I fear the email goes to me [06:36] as I signed the release [06:37] I just got another rejected message [06:37] ok, afk again, I will try to manage the shower now [06:37] via email? [06:37] and double check my bag [06:37] yes [06:53] mborzecki: hey (late reply) - no, Multipass does not plug removable media, what's the use case? [06:54] Saviq: hey, a scenario when you build a snap using snapcraft & mutipass, and your source tree is somewhere under /media, that would not work currently, would it? [06:55] mborzecki: it should, we're re-confining the mount process [06:58] mborzecki: ah hmm, but the `multipass` app will barf on trying to mount b/c of ENOPERM [06:58] Saviq: that's what i was suspecting, but i'm trying to mount something locally now [07:12] mborzecki: https://github.com/canonical/multipass/issues/1598 btw [07:13] Saviq: btw. can you try to mount something, unmount and mount the same path again? does it work for you? [07:20] Saviq: https://paste.ubuntu.com/p/vYptW9MPPB/ pretty sure it worked once, i could see content under ~/Home in the vm [07:54] mborzecki: worked just fine [07:55] nothing in your log suggests it didn't… [08:15] Saviq: hmm for some reason ~/Home was empty, i can try a look into that later [08:21] PR snapd#9069 opened: tests/main/selinux-clean: workaround SELinux denials triggered by linger setup on Centos8 (2.45) [08:43] the tumbleweed images are broken apparently [08:44] 2020-07-27T21:56:14.0598453Z 2020-07-27 21:56:14 Server google:opensuse-tumbleweed-64 (jul272153-296516) is taking a while to boot... [08:44] 2020-07-27T21:57:13.1611691Z 2020-07-27 21:57:13 Cannot allocate google:opensuse-tumbleweed-64: cannot allocate new Google server google:opensuse-tumbleweed-64 (jul272153-296356): cannot find ready marker in console output for google:opensuse-tumbleweed-64 (jul272153-296356) [10:45] mborzecki: I approved https://github.com/snapcore/snapd/pull/9019, but it seems it needs a merge from master to pick up some fixes [10:45] PR #9019: tests/nested/minimal/minimal-smoke: run core smoke tests in a VM meeting minimal requirements [10:45] ijohnson: thanks, let me update it now [10:48] ijohnson: hi, I reviewed #9063 [10:48] PR #9063: cmd/snap/debug: add "snap debug seeding" command for preseeding debugging (4/N) [10:48] hi pedronis thanks, let me take a look [10:59] ijohnson: also #9067 [10:59] PR #9067: cmd/snap-bootstrap/initramfs-mounts: add doSystemdMount + unit tests [11:24] ijohnson: I answered to some of mborzecki comments to #9063 with my input [11:24] PR #9063: cmd/snap/debug: add "snap debug seeding" command for preseeding debugging (4/N) [11:25] thanks pedronis makes sense [11:31] pedronis: I wonder if we should have the api for debug seeding use *time.Time in the response, that would make the JSON not include things like `0001-01-01T00:00:00Z` for some cases, it would slightly complicate the logic to test for nil, but maybe that would be good [11:32] ijohnson: I thought of that, and is a bit cleaner, otoh it's a debug api, but if you want to do that, it's ok with me [11:32] I was just looking at the case for classic system that was not preseeded, and the time result is very confusing actually, 2562047h47m16.855s is actually like 292 years so I'm a bit confused how we got that result [11:33] I would have expected in that case an output like `seeding-time: -` because we don't actually have in the state.json how long seeding took [11:33] we have seed-time [11:33] but not seed-start-time [11:34] there might be a few more cases where the client code needs to ignore values [11:34] sorry when I said `seeding-time: -` I meant `seeding-completion: -` [11:35] so what would you expect `seed-completion` to show in that case where we don't have a seed-start-time ? [11:35] ijohnson: yes, my point is, we have had seed-time since long time [11:35] I mean at the end of the day this is a debug command [11:35] if it shows weird things on unsupported scenarios it's probably ok [11:36] yes, otherwise I would say either ? or - [11:36] or we need to think if we can guess a start-seed-time but it gets weird [11:36] I will timebox it, if I can fix it in 10 minutes I will fix it, otherwise classic / non-preseeded systems will just be buggy with this command [11:37] why classic? [11:37] you just mean older systems without all the values, no? [11:38] yeah sorry that's what I mean [11:38] ah ok [11:38] I just said classic cause I can test this case easily on my classic system [11:38] BTW when was SeedTime added ? [11:38] err [11:38] seed-time [11:38] I would need to check [11:39] was that a long long time ago sufficient that I can probably assume it will always be there ? [11:40] ijohnson: I think it was added when we added the 2h delay for classic [11:40] but it will take me a bit to confirm because the code has been moved around [11:40] so blame doesn't give an immediate result [11:41] that's ok, I will assume it's always there and if it's not I think the code will just do "-" anyways I think [11:45] ijohnson: Mar 2018 [11:45] so bionic times [11:45] mmm seems old enough to me to say that it will always be there [11:45] for a debug command at least [11:45] yes [12:01] ijohnson: https://github.com/snapcore/snapd/pull/9064 spread jobs are close to finishing, we should restart it once more to make sure that caching of snapd snap job works [12:01] PR #9064: .github/workflows: move snap building to test.yaml as separate cached job [12:02] ijohnson: and btw the artifact is available already [12:03] mmm nice [12:03] so probably it shows up after the first run, and if you restart it, it doesn't discard that one from the UI [12:04] pedronis: https://github.com/snapcore/snapd/pull/9066 i think this one can be merged now, the failure on arch is unrelated (actually got 408 from the store) [12:04] PR #9066: o/ifacestate: fix bug in snapsWithSecurityProfiles [12:08] ok, made all the local changes, seems it took a bit longer than 10 minutes but meh [12:34] PR snapcraft#3232 opened: build providers: install apt-transport-https [12:54] cachio: hey, tumbleweed images seem to be broken, were they updated recently? [12:54] pedronis: #9067 is updated with your feedback addressed [12:54] PR #9067: cmd/snap-bootstrap/initramfs-mounts: add doSystemdMount + unit tests [12:55] mborzecki, hey, yes, yesterday [12:55] do you have a link to see the error? [12:55] cachio: 2020-07-28 11:10:01 Cannot allocate google:opensuse-tumbleweed-64: cannot allocate new Google server google:opensuse-tumbleweed-64 (jul280906-365867): cannot find ready marker in console output for google:opensuse-tumbleweed-64 (jul280906-365867) [12:55] this is all i get [12:55] cachio: looks like they are not booting at all? [12:56] yes [12:56] I can revert the image [12:56] mborzecki: did you restart spread on #9064 ? if so then the caching didn't work [12:56] PR #9064: .github/workflows: move snap building to test.yaml as separate cached job [12:57] ijohnson: no, i haven't [12:57] mborzecki, let me take a look first [12:57] mborzecki: ah ok [12:57] mborzecki: I will re-run spread after the current round of runs finishes [12:57] to ensure that the caching works [12:57] ijohnson: ok [12:57] hope the queue is not stuck again [12:58] mborzecki: no all the jobs are in progress, I just saw you approved, and wasn't clear if you approved because you saw the cacching work after restarting it or not [12:58] mborzecki: thx, merged, I will prepare a backport a bit later [13:01] mborzecki, I see 2020-07-28T13:00:33.756313+00:00 localhost systemd-vconsole-setup[3050]: syntax error, unexpected ERROR [13:01] 2020-07-28T13:00:33.758590+00:00 localhost systemd-vconsole-setup[3049]: /usr/bin/loadkeys failed with exit status 1. [13:01] cachio: that's on the console with updated image? [13:02] mborzecki, yes [13:03] PR snapd#9066 closed: o/ifacestate: fix bug in snapsWithSecurityProfiles [13:13] Hi. [13:13] Why won't you fix your mailserver? [13:31] quick errand to pick up the pi4 [14:07] re [14:28] geee restarted the spread job in #9006 since the build status was showing like half of jobs queued [14:28] PR #9006: bootloader: compose command line with mode and extra arguments [14:30] now after restart, most of them are complete, and it's not the first time, as if there's some issue reporting job status [15:01] cachio: fwiw the tumbleweed spread jobs are running again, have you reverted the image? [15:10] mborzecki, yes [15:11] cachio: cool, thank you! [15:11] I am creating a new image to test [15:11] mborzecki, yaw [16:19] ijohnson, cmatsuoka hey [16:19] I see hte new logs https://paste.ubuntu.com/p/F5HvDdvvnW/ [16:19] and the reboots are related to this error ide_atapi_cmd_error [16:19] look at line 1578 [16:20] also I see this: vmport: unknown command 56 [16:23] cachio: ah that's really interesting I bet that's related [16:32] * ijohnson afk for break [16:36] * cachio lunch [17:01] * ijohnson considers getting a specific boot for kicking github actions more [17:29] PR snapd#9070 opened: o/ifacestate: fix bug in snapsWithSecurityProfiles (2.45) [17:30] cachio: ah interesting [17:31] mborzecki: ijohnson: backport ^ [17:34] ijohnson: mmh, #9070 is running a differen set of tests? [17:34] PR #9070: o/ifacestate: fix bug in snapsWithSecurityProfiles (2.45) [17:34] pedronis: hmm ? [17:34] * ijohnson looks [17:34] woah what [17:35] maybe I'm just confused, and some haven't been added yet? [17:35] we haven't pushed changes to actions to 2.45 [17:36] also Maciej's previous PR today had all the tests [17:36] I'd say let's just wait to see what happens [17:37] this looks fine: https://github.com/snapcore/snapd/pull/9069 [17:37] PR #9069: tests/main/selinux-clean: workaround SELinux denials triggered by linger setup on Centos8 (2.45) [17:39] cmatsuoka, I am running now with more specific traces [17:39] to see if I can find something else [17:40] cachio: nice, I'm checking that unknown command to see what that could be [17:46] pedronis: ah ok, so after the unit test runs passed, I see all the spread ones show up now too [17:46] ok, just me confused [17:46] good [17:46] pedronis: so I think the UI is just slow to show them all [17:47] pedronis: it might have something to do with those checks not being required for the release branch [17:47] I dunno though [17:59] PR snapd#8677 closed: cmd/snap, daemon: detect and bail purge on multi-snap [18:08] ijohnson: can you look at #9068, I made a comment but not sure it makes sense or not [18:08] PR #9068: tests/main/selinux-clean: workaround SELinux denials triggered by linger setup on Centos8 [18:08] sure I can look but I'm no SELinux expert [18:10] thanks for the reviews pedronis [18:20] cmatsuoka, I reproduced this as well [18:20] https://paste.ubuntu.com/p/FcBqbvC4jj/ === pedronis_ is now known as pedronis [18:30] PR snapcraft#3222 closed: fix typo [18:30] PR snapcraft#3223 closed: sentry: don't report tool missing errors [18:31] cachio: can you resolve the conflicts in https://github.com/snapcore/snapd/pull/8558? [18:31] PR #8558: tests: make the nested library usable independently of spread [18:32] cachio: I think the log is not in the last paste [18:32] it's one level of indirection away [18:32] ijohnson, I think I'll close that one and open this in small ones [18:33] cachio: ack sounds goof [18:33] *good [18:35] PR snapcraft#3171 closed: snap: debug enabled by default [18:37] cachio: could you try with -machine q35,vmport=off [18:39] cachio: the unknown vmport command should be only a debug message, it doesn't seem that the crash is related to it but that property disables vmport [18:43] surew [18:43] cmatsuoka, running [18:51] cachio: my expectation is that it will still crash, but you won't see the vmport message in logs [18:54] PR snapd#9069 closed: tests/main/selinux-clean: workaround SELinux denials triggered by linger setup on Centos8 (2.45) [18:56] cachio: these are the known vmware i/o port commands, they go from 0 to 43: https://sites.google.com/site/chitchatvmback/backdoor [18:57] checking [19:00] PR snapcraft#3147 closed: Use `pylxd` instead of `lxc exec` [19:00] PR snapcraft#3156 closed: log: trace HTTP connections with developer debug [19:15] PR snapcraft#2890 closed: extensions: add opengl extension to support classic and strict [19:16] cmatsuoka, well, with the vmport=off the command error is not shown anymore [19:20] PR snapcraft#2413 closed: kernel plugin: introduce a _get_kernel_version() helper [19:22] cmatsuoka, do you have all the parameters for -machine? [19:39] PR snapd#9070 closed: o/ifacestate: fix bug in snapsWithSecurityProfiles (2.45) [20:03] cachio: I used the qemu-system(1) man page to see the parameter list [20:04] cmatsuoka, I am testing some specific parameters for intel machjnes [20:05] cachio: did you get any new logs when the vm crashes? [20:06] cmatsuoka, yes, but all is similar [20:06] ah ok [20:15] PR snapd#9071 opened: release: 2.45.3.1 [20:15] PR snapd#9072 opened: packaging: add placeholder changelog for 2.45.3.1 [20:15] ijohnson: I re-prepared the release ^ [22:05] PR snapd#8499 closed: Adding comment on system instability caused by a privileged cp