[00:45] PR snapcraft#3455 opened: python v2 plugin: filter set options to reduce output noise [00:55] PR snapcraft#3456 opened: python v2 plugin: reduce noise by replacing for-loop with xargs [03:05] PR snapd#9975 opened: tests: fix new tumbleweed image [07:01] morning [07:04] morning [07:22] morning [07:22] good morning [07:22] * zyga-mbp will make a snapd patch today [07:39] hey zyga-mbp [07:40] hey pstolowski :) [07:54] mvo good morning [07:54] mvo do you remember the nfs home and special workarounds [07:54] there's another case of that with nsswitch and ldap and home directory being defined by a ldap entry [07:55] mvo would it be okay to enable the nfs workaround when nsswitch has an ldap entry? [07:55] I can provide the patch [08:01] PR snapd#9957 closed: ovelord/snapshotstate: keep a few of the last line tar prints before failing [08:03] zyga-mbp: oh, interessting. what exactly was it we did for nfs homes? I vaguely remember [08:03] zyga-mbp: fwiw, I think it makes sense, without this the user can't use snaps at all, yes? [08:03] mvo: there are two components, we use some heuristics to detect that nfs home is likely to be used by looking at /etc/fstab and the mount table [08:03] mvo and when that is detected we update the apparmor profile of all snaps and helper tools [08:04] mvo: without this they get network errors on all snaps, apparmor has a bug where it exposes the fact that something uses nfs as network traffic on behalf of the process [08:04] this is bug https://bugs.launchpad.net/snapd/+bug/1917348 [08:04] Bug #1917348: NFS access not permitted for snap's on LDAP autofs system [08:05] I wanted to chat before spending more time on it [08:08] zyga-mbp: iirc we did some workarounds in the s-c profile? is this you're referring to? [08:08] mvo: hey [08:08] it's both snap-confine and actual snaps [08:09] the workaround is okay and doesn't need changes [08:09] only the heuristics needs to grow to check for ldap [08:09] hey mborzecki [08:09] zyga-mbp: yeah, I think the reporter has a point [08:11] mvo okay, I'll send a patch either tonight or tomorrow [08:11] thanks! [08:12] thank you! [08:18] mvo: can you merge #9962, the failure on 20.04 is unrelated? [08:18] PR #9962: asserts: include the assertion timestamp in error message when outside of signing key validity range [08:18] mborzecki: sure [08:18] thanks! [08:21] PR snapd#9962 closed: asserts: include the assertion timestamp in error message when outside of signing key validity range [08:21] PR snapd#9972 closed: daemon: move the last api_foo_test.go to daemon_test === pedronis_ is now known as pedronis [08:33] another day, another small spread patch [08:34] niemeyer_ do you have a moment to look at https://github.com/snapcore/spread/pull/112 [08:34] PR spread#112: Bump base version of go to 1.13 [08:34] niemeyer_ it will unblock some other patches [09:01] PR snapd#9976 opened: tests/unit/go: use tests.session wrapper for running tests as a user [09:28] mvo: little tweaks in https://github.com/snapcore/snapd/pull/9970#pullrequestreview-601601464 [09:28] PR #9970: packaging: drop 99-snapd.conf via dpkg-maintscript-helper [10:01] PR snapd#9977 opened: bootloader/lkenv: add recovery systems related variables [10:34] pedronis: #9930 is green, any preference for squash-merging or not? [10:34] PR #9930: asserts: pool changes for validation-sets [11:05] pstolowski: as you prefer, if you think history is too messy I'm fine if you want to squash-merge it [11:08] pedronis: ok thanks [11:12] PR snapd#9930 closed: asserts: pool changes for validation-sets [12:25] pstolowski: did you squash the other one? maybe now you need to rebase the follow-ups [12:26] pedronis: i didn't, you're right i probably need to rebase -i [12:45] pedronis: it turned out too messy, i reopened as #9978 [12:45] PR #9978: asserts: use Fetcher in AddSequenceToUpdate [12:46] PR snapcraft#3456 closed: python v2 plugin: reduce noise by replacing for-loop with xargs [12:46] pstolowski: ok [12:47] PR snapd#9964 closed: asserts: use Fetcher in AddSequenceToUpdate [12:47] PR snapd#9978 opened: asserts: use Fetcher in AddSequenceToUpdate [12:47] pstolowski: would it be too hard to do two commits? one with the other changes, and one with the new functionality [12:48] pedronis: probably not, let me try [12:48] thx [13:07] PR snapd#9942 closed: boot: helper for checking and marking tried recovery system status from initramfs [13:10] pedronis: done [13:12] nb. git add -p is pretty useful for that in case that helps anyone in the future [13:12] pedronis: now that 9942 landed, i've pushed an update to #9940 and it needs reviews [13:12] PR #9940: boot: cmd/snap-bootstrap: handle a candidate recovery system v2 [13:12] pstolowski: magit ftw ;) [13:14] :) [13:40] mborzecki: ok, what's the state of #9921, I forgot? [13:40] PR #9921: boot: helper for setting up a try recover system [13:48] pedronis: i've just pushed an update there [14:21] PR snapcraft#3457 opened: porting conda plugin from v1 to v2 so we can use it in core20 [14:31] PR snapcraft#3455 closed: python v2 plugin: filter set options to reduce output noise [14:46] pstolowski: I approved #9978 [14:46] PR #9978: asserts: use Fetcher in AddSequenceToUpdate [14:49] * cachio lunch [15:12] PR snapd#9910 closed: client, o/servicestate: expose enabled state of user daemons [15:37] PR snapd#9971 closed: daemon,o/c/configcore: introduce users.create.automatic [15:43] ijohnson: thank you! [16:46] pstolowski: I made a code org comment in #9559, if I remember correctly how things were meant to be structured [16:46] PR #9559: client, daemon, cmd/snap: cleanups from #9489 + more unit tests [16:46] you ok mup ? [16:46] that pr was merged 4 months ago [16:47] oh I see must have been wrong pr number from pedronis comment [16:47] yes, I meant #9959 [16:47] PR #9959: o/snapstate: update validation sets assertions with auto-refresh [16:47] pedronis: thanks [16:47] sorry [16:48] no worries, I just got a notification about a mention from mup and was very confused since it says I merged that PR and I didn't recall merging any pr's today [16:53] PR snapd#9979 opened: packaging/ubuntu-16.04/rules: turn modules off explicitly <âš  Critical> [16:55] ^ should fix hirsute spread test runs, which are currently entirely broken [17:08] PR snapd#9980 opened: o/devicestate: test that users.create.automatic is configured early [17:23] PR snapd#9981 opened: gadget: policy for gadget/kernel refreshes [17:23] pedronis: fwiw, I pushed a draft with the refresh policy for gadget/kernel (9981), needs the second part still but let me know if it heads into the wrong direction or if I misunderstood anything. there is a comment in update.go with the checks we need to do [17:29] mvo: I looked at the comment (only) and it matches what we discussed === ijohnson is now known as ijohnson|lunch [19:25] pedronis: 9981 now has the gadget part of the policy and a manager test that tests going from the old to the new world end-to-end. still some unit tests missing but it should be ok for a first look, especially the manager test. the first auto-refresh will error on the kernel but the subsequent auto-refresh will pull it in [19:26] pedronis: anyway, not urgent but some progress at least :) [19:53] PR snapd#9967 closed: tests: use a pre built govendor dependencies for opensuse tumbleweed <â›” Blocked> [20:00] * cachio afk === ijohnson|lunch is now known as ijohnson [21:44] PR snapd#9983 opened: tests: use new commands of os.query tool on tests [21:53] PR snapcraft#3458 opened: requirements: pip freeze [21:58] PR snapcraft#3459 opened: repo: account for arch & version when filtering stage packages [22:49] PR snapd#9984 opened: boot/flags.go: add SetNextBootFlags, NextBootFlags for snapd_next_boot_flags bootenv