[05:17] morning [05:35] hi! [05:45] mardy: heya [07:09] morning [07:15] good morning [07:19] good morning pstolowski and zyga [07:19] mvo, I've sent some fixes for spread [07:19] mvo, mainly go vet stuff for now [07:19] zyga: yeah, just vaw it [07:19] saw [07:20] mvo, https://github.com/snapcore/spread/pull/117 is a real tiny bug [07:20] mvo, some more in humbox [07:20] mvo, all of this is from go vet reporting things [07:21] zyga: looking [07:21] mvo, I've also started looking at what it takes to fix https://github.com/snapcore/spread/issues/119 [07:22] mvo, I'll try to enable travis-ci.com on my fork first [07:22] zyga: we have https://github.com/snapcore/spread/pull/107 that would also work, yes? [07:23] zyga: or is there a reason to stay with travis? [07:23] ah, I missed that, yes [07:24] mvo, older but relevant fix is also here https://github.com/snapcore/spread/pull/111 [07:24] zyga: looking [07:25] mvo: zyga: pstolowski: morning guys [07:25] mborzecki, hey :) [07:26] * zyga is still sick, so so-so morning ;) [07:27] mborzecki: good morning [07:27] zyga: uh, get well! [07:27] mvo: I added a new section with some ideas in the mount-control document, it would be nice if you had a look [07:28] oh, and hi mvo, zyga, pstolowski :-) [07:29] hey mardy [07:34] good morning mardy ! let me check [07:35] zyga: I think I will mark a bunch of yours as "ready" and see if I can have another review meeting with gustavo [07:35] mardy: this is the "Suggestion for generating mount units" section ? [07:38] mvo: yes [07:40] mvo, thank you! [07:46] mvo: BTW, I think that the failures in https://github.com/snapcore/snapd/pull/10375 are not due to my changes (maybe you can double-check), so please consider whether you should use your magic powers on it [07:47] mardy: sure, looking [07:48] mvo: but if you merge, please squash :-) [07:49] mardy: the 14.04 test failure is real, there is no systemd-run on 14.04 so the fake store setup will not work, I think it's okay to exclude 14.04 [07:50] mardy: i.e. just something like "systems: [-ubuntu-14.04-*] [07:50] mardy: but let me look at the others too [07:51] mardy: I can also do this if you want [07:54] mardy: the debian-10 failure also looks real, I don't remember the detail here though, i.e. where/why we check for the right libseccomp [07:55] mardy: and on all of the cores I think theere is a failure in "userdel -f snap_microk8s" - I think "userdel --extrausers -f snap_microk8s" is needed here [07:56] mvo: let me have a look [08:06] mvo: can you take a look at https://github.com/snapcore/snapd/pull/10481 ? [08:15] mborzecki: added it to my list :) [08:15] mborzecki: not sure if I manage before lunch though :/ [08:19] mvo: sure [08:19] thanks! [09:41] mvo: pstolowski: also a very simple PR https://github.com/snapcore/snapd/pull/10489 [09:43] sure [10:21] sorry if I missed this, but are the spread failures on tests/main/refresh-delta known issues? [10:32] mardy: I don't think they are [10:32] mardy: but the 14.04 test is still unhappy :) [10:35] mardy: it looks like deltas were pruned this morning by the store [10:35] mardy: let's see if they come back [10:41] mardy: maybe you can push the fix for ubuntu-14.04 (just exclude it) [10:41] mardy: and then we will see if the store auto-regenerates the relevant deltas [10:47] mardy: just talked to the store and the failure to get the deltas is being investigated (and it seems also understood) [12:28] mardy: i think you reviewed the other s-c PR, can you take a look at https://github.com/snapcore/snapd/pull/10492 ? [12:57] mborzecki: will continue right after the standup === axino` is now known as axino [13:54] deltas aren't available? i see a bunch of refresh-delta tests failing across all systems [13:57] mborzecki: they should be back === pedronis_ is now known as pedronis [14:24] ijohnson[m]: I reviewed https://github.com/snapcore/snapd/pull/10488 it needs some more work I think [15:02] pedronis: pushed a tweak to https://github.com/snapcore/snapd/pull/10469 [15:02] and thanks for the review! [15:07] mborzecki: are there real failures in https://github.com/snapcore/snapd/pull/10481 ? [15:09] pedronis: you mean the cloud config tests? they all seem to fail on acking the snap declarations signed by developer 1 [15:11] mborzecki: are you saying they fail on master too? [15:12] hm let's see if i can find a branch that fails that way too [15:13] if not i'll investigate, but it's unclear how snap ack can be influenced by those changes in boot [15:22] how do the files /var/lib/snapd/apparmor/profiles/snap-update-ns.* contribute to the apparmor profile? In other words, when are these rules loaded? [15:26] mardy: snap-confine runs snap-update-ns before running snap app, see also the description at the top of every snap-update-ns.* profile (Description: Allows snap-update-ns ....) [15:29] mardy: also look at the comments about snap-update-ns in snap-confine.c [15:29] pstolowski: mmm... so, this is the profile under which snap-update-ns will be run for the specific snap, and has little (if not nothing) to do with the profile of the snap itself? [15:39] mardy: yes i think so, its only use is to set up mount namespaces, maybe mborzecki can confirm [16:03] zyga: can you please review spread#116? would love to mark it "ready" for gustavo but for that it needs at least one +1 [16:06] mvo, looking [16:06] mvo, and *thank you* for asking :) [16:14] done [16:15] mvo, thank you for pushing tests on some of the refactors and fixes [16:16] mvo, I have some tests lined up but mainly for the all-new exporter logic [16:16] mvo, make sure to review https://github.com/snapcore/spread/pulls as it's one of the more serious problems reported by vet [16:28] * cachio lunch [17:13] cachio: i've addressed your comments https://github.com/snapcore/snapd/pull/10476, please take another look [17:15] pstolowski, sure [17:15] thanks [17:19] pstolowski, hey [17:19] why are you locating tests/lib/snaps/store/test-snapd-refresh-control-provider.v1/build-aux/snap/snapcraft.yaml there? [17:19] and not in the base directory of the snap? [17:21] cachio: i think build-aux is a newer approach [17:21] cachio: not strictly necessary [17:21] I use to locate it in tests/lib/snaps/store/test-snapd-refresh-control-provider.v1/snapcraft.yaml [17:22] if it works it is ok [17:22] yes it works [17:22] cachio: see https://snapcraft.io/docs/release-notes-snapcraft-3-3 [17:22] i've re-built all 4 snaps with snapcraft and re-uploaded them to the store [17:23] nice, so you can choose among the baswe dir and build-aux [17:23] makes sense [17:24] cachio: it's useful to keep main directory cleaner, of course not a problem here [17:24] pstolowski, right [17:26] pstolowski, just remember to ask store team to move the snaps to test-snaps-canonical user [17:26] cachio: ah, can they do this? [17:26] yes [17:26] nice [17:26] all the test snaps need to be reasigned [17:27] you will remain as a collaborator of the snap [17:30] cachio: cool. i'll talk to them tomorrow [17:30] thanks [17:31] pstolowski, also you need to set them as unlisted [17:39] cachio: can i do this myself? [17:41] in the store page, you go to edit -> settingd [17:41] settings [17:41] there you can set it as unlister [17:41] but I see there are tons of snaps not properly configured [17:41] cachio: ah, nice, thanks [17:43] ok, done [17:43] and eod, cu [17:49] ijohnson[m], hey, when you have a bit of time, could you please check https://github.com/snapcore/snapd/pull/10409 [17:49] I need to validate the idea because I changed a lot the PR [17:52] cachio sure I've got it on my queue now [17:55] tx [18:57] ah pedronis is gone already [18:58] ijohnson: fyi: https://paste.ubuntu.com/p/Dcfqhy2vwK/ something is broken in nested tests on master [18:58] bboozzoo: hmm [18:58] I'll take a look, thanks for the pointer [19:03] ijohnson: if you don't find anything obvious, i'll try to poke at it tomorrow morning [19:03] bboozzoo: ack, I seem to recall seeing some changes to the fakestore stuff recently, perhaps we broke it without running the nested tests [19:03] ijohnson: possible, we don't run them too often, not sure if they even get run when merging to master [19:04] anyways, i'm off 😉 shouldn't even be at the keyboard right now [19:04] 🙂 ttyl [19:29] mborzecki, hey [19:29] https://paste.ubuntu.com/p/9phCDkgqMM/ [19:29] it is related to refresh-delta test which fails currently [19:30] it says not updates available but the snap is refreshed [19:30] is anyone takins a look to that one currently? [19:31] ijohnson[m], ~ [19:32] cachio: that's from the refresh-delta test ? [19:32] from refresh-delta-from-core [19:33] both are failing [19:33] same error I think [19:34] ijohnson[m], i have an instance open now [19:36] cachio: what is the full failure output from spread ? [19:37] ijohnson[m], https://paste.ubuntu.com/p/KVHFY74dTT/ [19:39] cachio: is the `journalctl: unrecognized option '--sync'` error relevant at all ? or is that harmless [19:40] it is harmless [19:40] ijohnson[m], it should be [19:42] it is weird because the log says it cannot refresh the snap [19:42] hmm [19:43] ijohnson[m], https://paste.ubuntu.com/p/wpqgRCN2ZM/ [19:44] this is the part of the journal log which is doing the refresh [19:44] cachio: can you run on this instance `journalctl --no-pager -u snapd` ? the output you pasted is cut off [19:45] ijohnson[m], https://paste.ubuntu.com/p/9gJ5PwHN9G/ [19:49] cachio: yeah the store is still not returning deltas correctly, can you ping someone in the store channel about this? [19:49] mvo pinged them this morning and they said it was fixed, but it seems it is still not fixed [19:49] sure [19:49] ijohnson[m], thanks [19:51] np