=== popey0 is now known as popey [02:28] PR snapd#10147 closed: browser-support: Allow firefox to run correctly under wayland [05:41] morning [06:05] good morning [06:34] zyga: hey [06:34] mborzecki o/ [06:34] mborzecki I see snow outside [06:35] zyga: (un)lucky you [06:35] it's just rain here [07:02] morning! [07:12] pstolowski: hey, welcome back! [07:17] hey, glad to be back :) [07:17] * pstolowski needs to reboot [07:30] pstolowski: good morning! great that you are back [07:30] hey mvo ! [07:41] mvo: hey [09:25] PR snapd#10140 closed: boot, overlord/devicestate: consider gadget command lines when updating boot config [10:34] pedronis: fyi, https://github.com/snapcore/snapd/pull/9043 is updated now that the accessChecker implementations are on master [10:34] PR #9043: daemon: replace access control flags on commands with access checkers [10:34] so a smaller diff [10:35] PR snapd#10127 closed: daemon: add new accessChecker implementations [10:50] jamesh: thanks, great, I put it back in my queue [11:26] * pstolowski lunch [11:49] is anyone here? [12:07] Trixelit: there is [12:08] pedronis: i've opened https://github.com/snapcore/snapd/pull/10148 the managers tests grew a bit [12:08] PR #10148: overlord/devicestate, overlord/snapstate: add task for updating kernel command lines from gadget [12:10] PR snapd#10148 opened: overlord/devicestate, overlord/snapstate: add task for updating kernel command lines from gadget [12:10] mborzecki: that's most of what we discussed for .extra ? [12:10] yes [12:11] anyway let's see how it looks once 43 is landed [12:15] I'm confused, snap installed core version 2.37, refreshing it says there's no new updates but github has 2.42 [12:16] certbot won't work without the newest version but I don't know how to install it [12:23] Trixelit: 2.37 is very old, debian and ubuntu can reexec directly to a version from the core snap, but other distributions rely on their pacakge repositories [12:24] so what do I do? [12:25] snap version [12:25] snap 2.37.4-1+b1 [12:25] snapd 2.37.4-1+b1 [12:25] series 16 [12:25] deepin 20.2 [12:25] kernel 4.15.0-30deepin-generic [12:32] is there a way to update snapd from github? [12:37] Trixelit: no, why would you want that? (sorry i missed earlier context if there was any; snapd should refresh itself from time to time) [12:37] when i installed core, it installed at 2.37, but there are no updates available when i try to refresh it [12:38] and certbot doesn't work unless snapd is at its latest version [12:38] pstolowski [12:40] Trixelit: what does 'sudo snap refresh core' say? [12:40] snap "core" has no updates available pstolowski [12:41] Trixelit: and 'snap list'? [12:41] Name Version Rev Tracking Publisher Notes [12:41] certbot 1.14.0 1093 stable certbot-eff✓ classic [12:41] core 16-2.49.2 10958 stable canonical✓ core [12:41] (please pastebin if long) [12:41] core20 20210319 975 stable canonical✓ base [12:41] hello-world 6.4 29 stable canonical✓ - [12:45] hmm that's weird, you have latest core, so snapd should be at that version (2.49.2) === alan_g_ is now known as alan_g [12:52] pstolowski: it's deepin? maybe reexec isn't used there [12:52] maybe they have their own package builds or sth [12:53] makes no sense as to why certbot doesn't work if I have the latest core [12:53] I just hope i can get this working tonight [12:55] ou'll likely have to ask deepin to update the snapd package, seems it is used from the deepin archive there [12:55] *you'll [12:55] no reexec on deepin IIRC [12:55] it's easy to check [12:55] the logic is in snapd [12:56] but even if the distro patched it, it would need alignment to really work [12:56] right, its a deepin problem, they need to update [12:56] So pretty much, certbot will never work [12:56] unless deepin actually updates it [12:58] what are the symptoms of "certbot no working" ... you didnt tell us yet [12:58] iirc debian stable reexecs [12:59] from what i read deeping is based on debian stable, so unless they did something with package version or disabled reexec explicitly it shoudl reexec right? [12:59] * zyga looks [12:59] Trixelit: can you run `SNAPD_DEBUG=1 snap version` and paste the output? [13:07] Trixelit what does /etc/os-release look like on deepin? [13:07] @mor [13:07] ah [13:07] mborzecki 2021/04/14 23:07:08.334235 cmd_linux.go:70: DEBUG: re-exec not supported on distro "deepin" yet [13:07] error: unknown command "2.37.4-1+b1", see 'snap help' [13:07] zyga [13:07] PRETTY_NAME="Deepin 20.2" [13:07] NAME="Deepin" [13:07] VERSION_ID="20.2" [13:07] VERSION="20.2" [13:07] ID=Deepin [13:07] HOME_URL="https://www.deepin.org/" [13:07] BUG_REPORT_URL="https://bbs.deepin.org/" [13:07] there you go [13:08] there's no ID_LIKE=debian [13:08] so no reexec [13:08] you can try adding that [13:08] and restarting snapd [13:08] What is reexec [13:08] ID_LIKE="debian" [13:08] but no promises [13:08] where do I put that? [13:08] reexec is a snapd feature [13:08] where instead of waiting for deepin to do the update [13:08] you can get snapd from the core snap itself [13:08] (or from the snapd snap) [13:09] it requires some effort to support so it is not automatic [13:09] no updates available [13:09] *if* you want to try that then edit /etc/os-release, add that line and restart snapd [13:09] but no promises [13:09] I just tried that [13:09] snaps are all up-to-date [13:09] restart snapd.service [13:09] and run snap version [13:10] holy it updated [13:10] snap 2.49.2 [13:10] snapd 2.49.2 [13:10] series 16 [13:10] deepin 20.2 [13:10] kernel 4.15.0-30deepin-generic [13:10] Thanks zyga [13:10] Trixelit if it works do let us know [13:10] as in, if nothing breaks after a while [13:10] reboot and see if it's all good later [13:10] fingers crossed [13:10] ill try certbot in a few mins [13:40] pstolowski: I should look at the last commint in https://github.com/snapcore/snapd/pull/10053 right? [13:40] PR #10053: o/snapstate: helper for getting snaps affected by refresh, define new hook [13:41] pedronis: yes [13:41] ok, I put it back into my queue [13:55] PR snapd#10149 opened: many: implement fde-hooks v2 [14:28] thanks for the review mvo ! [14:28] my pleasure [14:30] PR snapd#10133 closed: wrappers/services.go: introduce EnsureSnapServices() [14:56] PR snapd#10150 opened: tests: new tests.skip tool [15:25] pstolowski: I reviewed https://github.com/snapcore/snapd/pull/10053, the code looks right but I was confused for a sec, made some comments/order suggestions [15:25] PR #10053: o/snapstate: helper for getting snaps affected by refresh, define new hook [15:26] pedronis: looking, thanks [15:56] PR snapd#10128 closed: tests: add 20.04 to systems for nested/core [16:00] jamesh: I re-reviewed https://github.com/snapcore/snapd/pull/8699 some small things there, I will need a 2nd review from somebody else in snapd [16:00] PR #8699: interfaces/desktop-launch: support confined snaps launching other snaps [16:00] s/I/it/ [16:00] * cachio_ lunch [16:48] re [16:48] hm, i'm confused by how we eanble secure boot in nested manual tests, iirc it was supposed by be enabled by default when the system is 20.04? [16:50] mborzecki, it always goes through this [16:50] [ "$NESTED_ENABLE_SECURE_BOOT" = true ] [16:50] because the var is set [16:51] what you changed is the default bahaviour in case the variable is not set [16:54] cachio_: i have nested/manual tests, which does not set NESTED_ENABLE_SECURE_BOOT, but system is 20.04, so secure boot would be enabled by default, wouldn't it? [16:55] eh and magically the same branch that works with tpm locally fails in spread jobs in 10148 [16:55] mborzecki, I think the mess was introduced when both core and core20 were merged [16:57] manual tests should use the same configuration as core suite [16:57] # Enable tpm in the nested vm in case it is supported [16:57] NESTED_ENABLE_TPM: '$(HOST: echo "${NESTED_ENABLE_TPM:-}")' [16:57] # Enable secure boot in the nested vm in case it is supported [16:57] NESTED_ENABLE_SECURE_BOOT: '$(HOST: echo "${NESTED_ENABLE_SECURE_BOOT:-}")' [16:57] so default values are used [16:59] mborzecki, do you want to update that in the same pr? [17:04] cachio_: i mean this part, the idea was taht if the env variable is not set, but the system is ubuntu-20.04* secure boot gets eanbled by default, and afaiu this covers tests/nested/manual tests too right? [17:05] uhh, i see now [17:05] there's env for tests/nested/manual suite which overrides this :/ [17:07] yes [17:07] this is what we need to change [17:26] PR snapd#10106 closed: secboot,boot: provide fde-hooks v2 API interface to hooks [17:27] * cachio_ afk [18:01] PR snapd#10143 closed: boot: handle updating of components that contribute to kernel command line [20:17] PR snapcraft#3500 closed: docker: Need to repeat ARG in every section [20:26] PR snapd#10118 closed: boot/flags.go: add HostUbuntuDataForMode [21:02] PR snapcraft#3502 closed: snaps: do not validate snaps before install/refresh (Fixes LP#1901733) [22:25] I can't run Chromium. Someone from #ubuntu suggested I try asking here. after running: sudo snap refresh, I got: error: cannot perform the following tasks: - Run configure hook of "chromium" snap if present (run hook "configure": cannot create temporary directory for the root file system: Permission denied) [22:37] StupidDanbo can you run `snap changes` and then `snap tasks ` for any task that is in Error state? [22:38] Also note that most folks are EOD right now so you will get the most help if you make a forum post and post some details then EU folks will see it in their morning [22:54] oh I only saw the 2nd part at first. [22:55] same error about permission denied for creating temporary directory for root [22:55] StupidDanbo can you share the full output though? [22:56] how do I do that? [22:56] You can copy the output from the terminal and put it in pastebin.ubuntu.com and share the pastebin link here [22:58] https://pastebin.ubuntu.com/p/s7GmChnnph/ [22:59] hm. identical error for task 270 [23:04] is there a way to backup my Chromium bookmarks before I try to reinstall it? or does"removing" all snaps not uninstall Chromium? [23:07] maybe "apparmor doesn't like symlinks" [23:08] but surely my /tmp/ isn't on the SSD... [23:08] .... oh, it IS a symlink. It's got that arrow on the folder icon. [23:09] ironically my root folder looks like it's on the SSD, while /tmp is pointing to my HDD to avoid filling it up. :|