[05:13] morning [05:17] Hey mborzecki [05:17] https://github.com/snapcore/snapd/pull/6937 [05:17] From last evening [05:17] PR #6937: cmd/snap-update-ns: detach unused mount points [05:17] hmhm what's that? :) [05:18] Embarrassment [05:20] jamesh: ^ if you can, can you have a look as well please? [06:01] I’m sleepy, will start late [06:16] zyga: will do. [06:30] Thank you :-) [06:38] zyga: reviewed, each time i look at this i actually have to go and take a look at the rest of the code around s-u-n [06:52] zyga: can you take a look at #6929 ? [06:52] PR #6929: gadget: record gadget root directory used during positioning [06:54] mborzecki: sure, back in 5 minutes [07:15] man, my timing sucks [07:15] back now [07:15] going to review [07:17] mborzecki: as for sun, do you have any ideas on how to make things clearer? [07:18] mborzecki: +1 [07:18] I think you can merge that as is [07:20] eh, retry branch is red [07:50] zyga: retry the retry branch? :) [07:50] no, it's really broken [07:51] maybe pstolowski will have some ideas [07:51] it feels like chasing a rabbit now [07:52] zyga: it has always been whack-a-mole thing [07:52] now it's whack some more or disable tests :/ [07:52] zyga: i can look at this later/monday (i'm going to take half-day off today) [07:53] thanks! [07:53] i don't expect this to be a quick fix, it's always painful [08:11] hey Chipaca [08:11] zyga: hiya :) [08:13] Chipaca: wanna do a 2nd review https://github.com/snapcore/snapd/pull/6937 ? [08:13] PR #6937: cmd/snap-update-ns: detach unused mount points [08:13] pretty short [08:14] Chipaca: morning sir! [08:16] switched over to firefox last night, looking good so far [08:16] Chipaca: from chrome? [08:16] yes [08:16] Chipaca: I play with firefox once in a while, it's pretty speedy! [08:16] only issue is with passwords that didn't auto-migrate [08:16] it's always surprising when I open it [08:16] ouch [08:16] but passwords.google.com to the rescue :-) [08:17] yeah, using safari on ios and macos makes me want to keep using them over other browsers [08:17] I'd switch easier if firefox integrated with the system keyring [08:17] and apparently it has a separate one :/ [08:17] and with that i discovered parts of aws leaking into amazon.com [08:18] (if you go to amazon.co.uk (or .com)'s "your account" → "login & security", the MFA it asks for is your _AWS_ MFA, which is separate from your amazon.co{m,.uk} MFA [08:18] ) [08:19] and, guess who still had their old phone and address in AWS (because I haven't used it since I moved)? [08:19] fun night [08:20] anyway, firefox is nice, and the whole "yeah we're breaking the good ad blockers, but hey at least we're dropping text adds altogether" google thing was too far [08:20] ads* [08:38] hey Chipaca [08:39] Chipaca: also switched to firefox from chrome recently [08:39] we should move our standup away from hangouts :-) [08:40] I need reviews for bugfixes: https://github.com/snapcore/snapd/pull/6937 and https://github.com/snapcore/snapd/pull/6891 [08:40] PR #6937: cmd/snap-update-ns: detach unused mount points [08:40] PR #6891: many: make per-snap mount namespace MS_SHARED [08:46] zyga: “To counter that, the writable mimic [08:46] construction code in snap-update-ns switched the stash mount point to [08:46] recursively private, so that prior semantics is retained.” [08:47] zyga: is that "switched" actually a "switches"? [08:47] (i have not looked at the code -- i'm asking if i should expect to find that switch in this pr) [08:47] ((and if not in this pr, where?)) [09:05] mborzecki: i wonder what "armound" comes from [09:05] amount maybe? [09:06] Chipaca: probably closest hamming distance [09:06] mborzecki: want to change the suggestion to an actual suggestion so we can commit it? or should I :) [09:08] Chipaca: go ahead :P [09:09] would be nice to know which syscalls are the problematic ones [09:09] so we can fix them in go itself =) [09:11] huh, i don't have permission to commit my suggestions [09:11] mborzecki: can you batch up the two suggestions and try to commit them? [09:12] (the other one is a trivial it's → its) [09:12] Chipaca: sure, let me see [09:13] taw [09:15] and pushed [09:15] taw² [09:16] funny git push is ok, but commiting on review page is not, though supposedly it's the same user attempting the action [09:17] ah, i was wondering :-) [09:27] Chipaca: can you take a look at #6929 ? [09:27] PR #6929: gadget: record gadget root directory used during positioning [09:29] mborzecki: yes [09:30] Chipaca: thanks! [09:32] editing distance is fun and i hate you for sending me down that fun rabbit hole [09:32] amount is editing distance 3, and there are several words closer [09:33] actually, just two: around and arround are distance 1. Nothing at distance 2. [09:35] Chipaca: arround? [09:35] using the british-english-insane wordlist, yes [09:36] Chipaca: not aground? [09:37] strangely no [09:37] but my code might be buggy [09:37] :) [09:37] that should be there with edit distance 2 though [09:37] hmm [09:37] * Chipaca closs the file and walks away [09:45] zyga: i'll be able to reproduce retry issue even without your PR when using unshared network ns? [09:45] PR snapd#6929 closed: gadget: record gadget root directory used during positioning [09:50] mborzecki: ineed i was stomping on distance n things with distance m>n things when they were the same but found later [09:50] mborzecki: https://paste.ubuntu.com/p/bMZDf67zHy/ [09:50] * Chipaca now yes deletes the file [09:51] Chipaca: haha :) feel genuinely sorry for triggering this [09:51] :) [09:55] pstolowski: re, partially, there are different failures with and without the patch [09:55] pstolowski: but yeah, just look at what the spread test does [09:55] zyga: ack, i've just reproduced httputil failure [09:55] Chipaca: I meant that the code that does handle the mimic now uses private sharing [09:56] zyga: where 'now' means 'as of this PR'? [09:56] yes [10:04] mborzecki: can you complete your review of https://github.com/snapcore/snapd/pull/6937 -- not sure if I need to change anything or just merge it [10:04] PR #6937: cmd/snap-update-ns: detach unused mount points [10:15] zyga: if a struct holds a pointer, copying the struct will copy the same pointer [10:17] I know that, [10:17] I was explaining why I change the options array [10:17] zyga: wrt the commit message, if you change "the writable mimic construction code in snap-update-ns switched [...]" to "this change switches the writable mimic code [...]", it becomes clearer [10:17] +1 [10:19] done [10:32] PR snapd#6937 closed: cmd/snap-update-ns: detach unused mount points [10:34] can we land https://github.com/snapcore/snapd/pull/6933 per my comment there? would be great to have it in edge after the weekend [10:34] PR #6933: [RFC] snapd: ensure GOMAXPROCS is at least 2 [10:42] PR snapd#6933 closed: [RFC] snapd: ensure GOMAXPROCS is at least 2 [10:48] +1 [11:02] * Chipaca takes a break [11:16] i'm off for today, have a great weekend guys, see you! [11:19] pstolowski: bye! [11:58] pstolowski, enjoy your weekend === epod is now known as luk3yx === davdunc_ is now known as davdunc === sdrobertw_ is now known as sdrobertw === JamieBennett_ is now known as JamieBennett === kalikiana_ is now known as kalikiana === Lukewh_ is now known as Lukewh === jamespage_ is now known as jamespage === benoitc_ is now known as benoitc === nottrobin_ is now known as nottrobin === joc__ is now known as joc === pstolowski_ is now known as pstolowski === moul_ is now known as moul === coreycb_ is now known as coreycb === davecore_ is now known as davecore === AndyWojo_ is now known as AndyWojo === ricab is now known as ricab|lunch [12:55] Snapcraft Live starts is a few minutes - https://www.youtube.com/watch?v=oR3XLnhypts === Greyztar- is now known as Greyztar === ricab|lunch is now known as ricab [13:56] /me lunch [15:04] * cachio lunch [15:06] * zyga had a glass of wine and is very dizzy now [17:55] Hello I wonder if it would be difficult to Snap "peerflix"? I mean to try this now but was just curious if I might be biting off more than I can chew? [17:55] I made a fork of "peerflix" here https://github.com/Paddy-NI/peerflix [17:55] I also want to snap "mps-youtube" [17:57] I have often found mps-youtube to only work if both it and youtube-dl are installed using "pip3". If you have installed youtube-dl via apt then it tends to break mps-youtube. [18:01] Paddy_NI: ooh, nice. If you're doing that, you could also patch libncurses to work with the mouse wheel :-) [18:01] (the one in ubuntu still doesn't) [18:02] anyway, EOW for me [18:02] 👋 [18:18] * cachio afk [19:04] I am trying to snap "peerflix" which succeeds of course when using "devmode" however when I switched to "strict" and snap installed it with "dangerous" it poops itself. Here is the output from "snappy-debug.security scanlog" https://paste.ubuntu.com/p/KhSdP2Bsp6/ [19:36] Hey popey it would be lovely if you shared your bashrc with us not locals sometime so we could all look at the precious things of the shop! [20:36] \o/ writable on tmpfs runs up to console-conf! [20:37] (of course console-conf won't do much without a real on-disk writable -- but that's the next step)