[02:58] PR snapd#11331 closed: gadget: mv modelCharateristics to gadgettest.ModelCharacteristics === mwhudson_ is now known as mwhudson [06:47] morning [06:48] PR snapd#11336 closed: cmd/snap: add --debug to snap run [07:17] hi! [07:33] mardy: heyaa [08:02] morning [08:04] PR snapd#11303 closed: overlord/servicestate: disallow mixing snaps and subgroups [08:59] PR snapd#11330 closed: spread: switch to CentOS 8 Stream image [10:09] PR snapd#11255 closed: many: transactional installs/removes/refreshes [10:09] PR snapd#11290 closed: cmd/snap-confine: mount bpffs under /sys/fs/bpf if needed [10:54] PR snapd#11338 opened: asserts,cmd/snap-repair: support delegation when validating signatures [12:25] PR snapd#11327 closed: snap-confine: allow numbers in hook security tag [12:26] mborzecki, pstolowski: do you know better ways to do this? https://github.com/snapcore/snapd/pull/11321/commits/edca569444e93526dc587622de283567319175eb [12:26] PR #11321: overlord: fix issue with snapshot restoring on CentOS [12:26] I wanted to use o.Settle(), but it's incompatible with o.Loop(), which the test uses [12:32] mardy: heh, was abou to write that the test shuld probably call settle, but this one is odd [12:36] mardy, mborzecki yeah my first instinct would be to try to make it use settle() instead, or is it problematic in this test? [12:39] mardy: you can also just wait for chg.Ready() channel, not sure if it helps? [12:40] <-chg.Ready() would block until ready afair [12:40] mardy: you can try something like https://github.com/snapcore/snapd/pull/11321/commits/edca569444e93526dc587622de283567319175eb#r798522418 maybe? but that's just a refactor of the code, nothing special [12:40] PR #11321: overlord: fix issue with snapshot restoring on CentOS [12:41] nvm, <-chg.Ready() is already there [12:47] pstolowski: yes, but I don't only need for the change to be ready, it needs to be clean (the Ensure() function goes through completed tasks and invokes their cleanup handler, or directly marks them clean if there is no handler) [12:48] mardy: yes i know, i assume Ready() is the first barrier; but then setting clean flag happens asynchronously as well [12:49] anyway, what mborzecki suggested is a more idiomatic way in go / a bit nicer [12:49] mborzecki: looks good, but I guess that I don't nee to to break back to the beginning of the loop; it seems to me that I can just use an ordinary "break" with no label [13:01] ah, no, my ignorance, nevermind :-) [13:55] PR snapd#11176 closed: tests/core/failover: verify failover handling with the kernel snap [16:54] zyga sorry to bother u again. could u have a look at https://github.com/snapcore/snapd/pull/11245 (vscode pr)? [16:54] PR #11245: vscode: added integrated support for MS VSCODE [16:56] PR snapd#11339 opened: o/servicestate: revert #11003 checking for memory cgroup being disabled <âš  Critical> [16:58] arsenique, re [16:58] arsenique, looking [16:58] thank u. [16:59] arsenique, I don't see any replies to any of the points I made about extra extensions [16:59] arsenique, did you send your replies? they can be stuck in github until you send the entire review response [16:59] https://github.com/snapcore/snapd/pull/11245#discussion_r795442967 [16:59] PR #11245: vscode: added integrated support for MS VSCODE [17:00] arsenique, after my comments there's a message from miguelpires but I don't see anything more recent [17:00] I only see my comment there [17:00] if you see more you must have a pending review [17:00] zyga look at my replies to your comments [17:00] there are no replies [17:00] check in a privater browser window [17:00] *private [17:00] please follow the link https://github.com/snapcore/snapd/pull/11245#discussion_r795442967 [17:01] again [17:01] the replies appear close to the comment, not as a new one [17:01] please open a private browser window [17:01] I don't see a single comment there [17:01] perhaps others can confirm [17:02] I'm 100% sure you have a pending review that you yourself can see [17:02] but until you send the review [17:02] those comments are invisible to others [17:02] yes I see it in all browsers I have [17:02] then github is broken [17:02] can you paste your comments? [17:03] (in all browsers that you are not logged in as yourself?) [17:03] arsenique zyga-office can confirm, I don't see any replies as well [17:03] try a private window, if that works [17:03] thanks miguelpires [17:03] arsenique, github UX sucks on this [17:03] you must be logged in [17:03] finish review and the replies will show up [17:05] did you try the link that I suggested? https://github.com/snapcore/snapd/pull/11245#discussion_r795442967 [17:05] PR #11245: vscode: added integrated support for MS VSCODE [17:06] can you share a screenshot? [17:07] yeap. [17:10] zyga sorry for a stupid question, is it possible to share screenshot here somehow? [17:10] arsenique, you have to use a 3rd party like imgurl [17:11] or twitter or whatever you prefer [17:11] got it [17:11] but irc itself is text only [17:11] arsenique: yes, I did. zyga-office is right, your comments are still pending (they say so at the top, next to your name). You made a review instead of replying so you need to finish the "review" by submitting [17:11] i feel so stupid [17:11] (for context, arsenique shared a screenshot with me OOB) [17:12] everyone been there ;) [17:12] github ui sucks [17:12] in this very thing at least...and big time [17:13] no worries arsenique [17:13] github is very broken on this aspect [17:13] you've waited for days without feedback [17:13] but I bet we're not unique here [17:15] thanks heaven gh is not loosing those pending comments [17:16] arsenique: did you submit the review? [17:17] I don't think so [17:18] funny part I am struggling to find this bloody review button. I am an author of this pr, so perhaps that is why I don't have this button. [17:18] or I am that thick :-( [17:19] you have it [17:19] arsenique: author has this button [17:19] anyone leaving a comment has two optinons [17:19] send the comment right now [17:19] or "start a review" [17:19] green "Review changes" in the "Fileschanged" tab [17:19] that other thing creates a stash of comments [17:19] what pstolowski said :) [17:20] I've been on gitlab for a while and I'm not used to github UI anymore [17:20] Yes, the green button at the top that says something like "Finish your review" [17:20] omg are they serious??????????????????????????????????????????????????????????? [17:21] usability 101 [17:21] do they know the meaning of ux? [17:21] seriously who is that genius that decided on how this flow shall work?! [17:22] I see your comments now [17:23] i am really puzzled right now. i don't know how to feel right now. being angry or stupid?! [17:24] zyga pstolowski miguelpires thank you all guys so much for your help and patience. [17:25] I replied to your comments now [17:25] arsenique, just let it out, it's very frustrating [17:25] I think everyone here bumped into this the same way [17:25] it's just meh [17:25] bad bad bad interface [17:27] thanks a lot guys. very much appreciated. [17:27] thanks for pushing forward :) [17:32] np [17:36] PR snapd#11340 opened: [systemd] fix systemd unit stop progress notifications [18:41] PR snapd#11341 opened: tests: updated the documentation to run spread tests using external backend [19:26] PR snapd#11342 opened: tests: add regression tests for disabled memory cgroup operation [20:07] PR snapd#11339 closed: o/servicestate: revert #11003 checking for memory cgroup being disabled <âš  Critical> [21:16] PR snapcraft#3631 closed: parts: add early base parsing (CRAFT-766) [21:27] PR snapd#11343 opened: tests: reboot test running remodel [21:42] PR snapd#11344 opened: cmd/snap/auto-import: use osutil.LoadMountInfo impl instead [22:12] PR snapd#11345 opened: tests/main/snap-system-key: reset-failed snapd and snapd.socket