/srv/irclogs.ubuntu.com/2019/04/17/#snappy.txt

mupPR snapd#6735 opened: tests: Wait for man db cache is updated before after install snapd on Fedora <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/6735>02:11
=== chihchun_afk is now known as chihchun
=== chihchun is now known as chihchun_afk
=== chihchun_afk is now known as chihchun
zygagood morning06:45
zygaHey mvo06:54
mvohey zyga06:58
pedroniszyga: mvo: hi07:00
zygaHey pedronis07:00
zygaI didn’t push it yet but I have the next chunk of update ns refactor rebased and adjusted07:00
zygaI’ll open the PR in about an hour after I handle kids and other morning routine07:01
mvohey pedronis - good morning07:07
pedroniszyga: ok, I likely will not get to it before my vacation though07:08
=== pstolowski|afk is now known as pstolowski
pstolowskimorning07:15
mupPR snapd#6659 closed: snapcraft: build static fontconfig in the snapd snap <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/6659>07:31
mvohttps://github.com/snapcore/core/pull/104 <- needs a review (should be simple :)07:32
mupPR core#104: snapcraft.yaml: use remote fc-cache-builder <Created by mvo5> <https://github.com/snapcore/core/pull/104>07:32
mborzeckimorning07:37
mvohey mborzecki07:37
mborzeckimvo: hey07:38
mborzeckia day off and i'm looking at spread logs :)07:38
mvomborzecki: I was wondering07:39
mvomborzecki: if I misremembered the day :)07:39
mborzeckimvo: got an email from travis that the job is still failing07:40
mborzeckilooks like google:centos-7-64:tests/main/nfs-support leaves garbage behind07:41
mvomborzecki: uh, ok - I saw this failing earlier indeed07:41
mupPR snapd#6418 closed: many: allow core as a fallback for core16  <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/6418>07:43
mupPR snapd#6603 closed:  snapstate: add new NoReRefresh flag and use in Remodel() <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/6603>07:44
mborzeckimvo: it's probably best to disable the test for fedora and centos until i figure out a way of this08:14
mborzeckimvo: in short, nfsv3 does not export selinux labels, nfs 4.2+ can08:15
mborzeckimvo: so with nfs3 incorrect label (nfs_t) will be used, regardless of what we do08:15
mborzeckimvo: just checked that nfsv4 with security_label option works correctly08:18
zygaquick trivial PR ahead of other stuff https://github.com/snapcore/snapd/pull/673608:24
mupPR #6736: cmd/snap-update-ns: rename variable "up" to "ctx" <Simple 😃> <Created by zyga> <https://github.com/snapcore/snapd/pull/6736>08:24
zygamborzecki: are you off or not :)08:24
zygamborzecki: sho, go and enjoy the sunlight :-)08:24
mborzeckihahah08:24
mupPR snapd#6736 opened: cmd/snap-update-ns: rename variable "up" to "ctx" <Simple 😃> <Created by zyga> <https://github.com/snapcore/snapd/pull/6736>08:24
mupPR snapd#6737 opened: tests/main/nfs-support: temporarily disable for Fedora and CentOS <SELinux> <Simple 😃> <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/6737>08:24
mborzeckizyga: mvo: ^^08:24
zygaapproved08:25
mborzeckididn't want to leave you with master broken again :P08:25
mborzeckiok, now i'm off, cu tomorrow08:25
zygaah08:26
zygahey Chipaca08:26
Chipacahi08:27
Chipacawhy do i suspect forum-person tried the equivalent of 'snap install snapd'08:28
Chipacabecause they heard the snap needed snapd08:28
zygaChipaca: perhaps, I was thinking the same thing08:29
zygahttps://github.com/golang/go/commit/97c4ad432743d74ee59648dee0db1b107c70183409:32
zygaCL09:32
zygachange list09:32
zygaperforce09:32
zygapeople hacking on go use perforce?09:32
* pstolowski runs a quick errand, afk for a bit09:40
Chipacazyga: googlespeak leaking09:42
mupPR snapd#6599 closed: snapstate,state: add TaskSet.AddAllWithEdges() and use in doUpdate <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/6599>09:54
pedronismvo: maybe this was already clear but I added a post-facto comment to https://github.com/snapcore/snapd/pull/641810:09
mupPR #6418: many: allow core as a fallback for core16  <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/6418>10:09
zygamvo: https://api.travis-ci.org/v3/job/521150301/log.txt is interesting10:24
zygamvo: no idea, perhaps something left over core16 behind?10:24
mvozyga: yeah, that sounds like it :/10:25
zygamvo: mvo I saved https://paste.ubuntu.com/p/GVwfVfTHVm/10:25
zygamvo: I'll restart the run10:25
mvook10:25
* Chipaca goes to get some exercise10:25
mvopedronis: thanks for adding this extra info to the core16-core pr. quick question: checkSnap() needs to get access to the new model to validate things like kernel/gadget for the new model. should I add model to checkSnap or rather "task" and we add the info about the new model to the task. or something else?10:30
pedronismvo: we probably need to chat, I have a plan for that but is involved  (we really need to know a new store and new model at various points)10:42
mvopedronis: is now good? how long will it take? I can also shelve it and work on something else first if you want to do the prereq work10:43
pedronismvo: we can chat quickly now10:44
mvopedronis: ok, I'm in the standup ho10:44
pstolowskire10:48
zygagah11:09
zygawhen master is read on rename, it's not a good day11:10
zygaportal test failure https://www.irccloud.com/pastebin/PKt1NTxz/11:10
zygapedronis: https://github.com/snapcore/snapd/compare/master...zyga:feature/user-mount-ns-20.9-split-3-of-n?expand=111:14
zygapedronis: at +800, -200 I think I will split it into 2-3 more branches11:15
zygabut this is now rebased and I think in good shape11:15
zygaand on top of that I have working persisted and updated per-user mount namespaces11:15
zygaI will need to spend some time to research jdstrand's questions but it's closer now11:15
pedroniszyga: cool, as I said but likely might not get to look at it this week11:15
zygayep11:15
zygaI'll return to other pending branches11:15
mupPR snapd#6737 closed: tests/main/nfs-support: temporarily disable for Fedora and CentOS <SELinux> <Simple 😃> <Created by bboozzoo> <Merged by zyga> <https://github.com/snapcore/snapd/pull/6737>11:16
zygatime to fix https://github.com/snapcore/snapd/pull/671711:16
mupPR #6717: snap: fix interface bindings on implicit hooks <Created by zyga> <https://github.com/snapcore/snapd/pull/6717>11:16
pedronismvo: btw as I said I'm ok with 6733 with the small tweaks I proposed11:18
mvopedronis: thanks!, will get to it after lunch11:20
mupPR snapd#6738 opened: tests: check for /snap/core16/current in core16-provided-by-core <Created by mvo5> <https://github.com/snapcore/snapd/pull/6738>11:20
=== epod is now known as luk3yx
=== chihchun is now known as chihchun_afk
* cachio afk11:41
pedroniszyga: do I see it correcly that #6681 will/might conflict with our own snap-confine groups related rework?11:46
mupPR #6681: many: support system-users for 'daemon' user <Complex> <Created by jdstrand> <https://github.com/snapcore/snapd/pull/6681>11:46
pedroniss/our own/your own/11:46
zygayes11:46
zygaI spoke with jamie about it and promised to work on all conflicts11:46
pedronisok11:46
zygahe was happy with either of us doing the conflict resolution though11:46
zygaso I don't anticipate problems11:46
zygaeh12:03
zygaso that go check thing where it eats all memory12:03
zygathat is not fixed?12:03
zygawhen it diffs recursive structures12:03
Chipacazyga: i thought it was a bug in the differ, that had been fixed?12:13
Chipacanot sure tho12:13
zygaapparently not enough12:13
Chipacamaciej knows more12:13
zygaI fixed it locally12:13
zygaby disabling that12:14
Chipacadon't disable the maciej! it's nice having him around12:22
zygapstolowski: can you please look at https://github.com/snapcore/snapd/pull/671712:22
zygatoo late! he's gone today12:22
mupPR #6717: snap: fix interface bindings on implicit hooks <Created by zyga> <https://github.com/snapcore/snapd/pull/6717>12:23
pstolowskizyga: will do12:23
pedronisjdstrand: hi, I did a first pass, getting familar with #668112:36
mupPR #6681: many: support system-users for 'daemon' user <Complex> <Created by jdstrand> <https://github.com/snapcore/snapd/pull/6681>12:36
pedronisjdstrand: what's the status of #5644  vs landing it for 2.39 ?12:45
mupPR #5644: interfaces: add audio-playback/audio-record and make pulseaudio manually connect <â›” Blocked> <Created by jdstrand> <https://github.com/snapcore/snapd/pull/5644>12:45
jdstrandpedronis: thanks! I'll take a look. other than that, there is the outstanding question of how to deal with systems that do not have the daemon user12:45
pedronisjdstrand: there is a list of those in the forum topic, right?12:46
jdstrandpedronis: I've been unable to do any snap decls. I've hoped to get to it since the agreement. I still hop to start this week12:46
pedronisjdstrand: ok12:46
jdstrandpedronis: the forum topic only lists solus as not having the LSB-required daemon user. that is the only one I am aware of12:47
jdstrandpedronis: I have not tested actual behavior yet if it doesn't exist. I know that the lookup for u:daemon and g:daemon will fail in policy generation. how that is exposed to the user I'm not sure12:48
pedronisjdstrand: I think we should fail somehow around checkAssumes on this12:49
jdstrandpedronis: so, I could try to do something with an implied assumes, an explicit assumes, fail earlier or implement a new backend to create it. maybe you have other ideas12:49
jdstrandpedronis: I was leaning towards something like that as well12:49
pedronisjdstrand: I think we start by failing with an implied assumes12:50
pedronisand see what feedback we get from users there12:50
jdstrandpedronis: ok, I'll work on that next12:50
pedronisand whether that pushes us, the packaging or the distro to creat the user12:50
jdstrandzyga: I had a conflict today. I already resolved it12:50
jdstrandpedronis: what is the timeframe for 2.39 btw?12:50
pedronisjdstrand: it's in the forum now12:51
pedronisjdstrand: https://forum.snapcraft.io/t/the-snapd-roadmap/197312:51
jdstrandpedronis: I marked the daemon user for that since, well, lots of people want it, roadmap, etc. it shouldn't be very risky since most of the code only hits if specifying system-users12:52
jdstrandok, cool. I see the topic (hadn't read email yet today)12:52
pedronisjdstrand: that's fine, next week I'm off but I would think if we clarify things,  mvo and zygmunt reviews would be enough12:52
pedronisjdstrand: I have a question about the snap-confine change, I'm either confused or the comment is or yet something else12:53
jdstrandpedronis: joe said it's my highest priority, so I'll stay on top of it12:53
jdstrandpedronis: I saw, I'll swing back to your questions a bit later12:54
mupPR snapd#6627 closed: devicestate: deal correctly with the "required" flag on Remodel <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/6627>12:59
=== ricab is now known as ricab|lunch
jdstrandpedronis: ok, responded. I'll focus on the implied assumes behavior and keep an eye on the PR for things you or others would like to see changed13:26
jdstrandmvo: hey, fyi https://bugs.launchpad.net/snapd/+bug/1825052 in case you didn't see it13:31
mupBug #1825052: seccomp argument filtering not working on Fedora with 2.38 and Debian with 2.37.4 <snapd:New> <https://launchpad.net/bugs/1825052>13:31
jdstrandmvo: also, I wasn't planning on looking at that. if you need me to, we'd need to chat with joe (note, I've not really worked on cross distro enablement)13:33
jdstrandmvo: I'd also like hello-world to be updated to accept hello-world.sh -c ... (not to mention, be rebuilt so it passes review-tools resquash since it was built before -no-fragments)13:39
jdstrandmvo: I can submit something for that (or upload); istr it was part of the shared account but then the shared account went away... please advise :)13:40
mvojdstrand: I can add you to hello world, one sec13:49
mvojdstrand: you should have an email13:49
pedronispstolowski: #6678 can land right?14:15
mupPR #6678: cmd/snap, api: use DebugGet for timings <Created by stolowski> <https://github.com/snapcore/snapd/pull/6678>14:15
pstolowskilooking14:15
pstolowskiyes14:18
mupPR snapd#6678 closed: cmd/snap, api: use DebugGet for timings <Created by stolowski> <Merged by stolowski> <https://github.com/snapcore/snapd/pull/6678>14:18
jdstrandmvo: thanks!14:31
jdstrandmvo: it's the little things in life :)14:31
=== ricab|lunch is now known as ricab
mvojdstrand: heh - my pleasure! thanks for updating this snap14:58
=== chihchun_afk is now known as chihchun
mupPR snapd#6736 closed: cmd/snap-update-ns: rename variable "up" to "ctx" <Simple 😃> <Created by zyga> <Merged by zyga> <https://github.com/snapcore/snapd/pull/6736>15:05
mvo6716 needs a second review15:06
mvopedronis: 6720 has a conflict now15:10
mupPR snapd#6739 opened: cmd/snap, store, image: support for cohorts in "snap download" <Created by chipaca> <https://github.com/snapcore/snapd/pull/6739>15:26
Chipacapedronis: ^ first cohorts bit (starting from an unlikely one, but a nice narrow vertical)15:26
pedronismvo: possibly yes15:27
Chipacanext up, create-cohort15:27
Chipacawith an alias of 'cohorate', obvs15:28
pedronisChipaca: thx15:28
* pedronis is doing fun things in/to snapstate atm15:28
Chipacanooo15:28
Chipaca:-)15:29
* cachio lunch15:29
Chipacaooh, oh, i forgot to add a bit to that15:29
* Chipaca force-pushes quickly before anybody notices15:29
jdstrandmvo: fyi, I never got an email15:35
* Chipaca quickly sends jdstrand his first email15:36
jdstrandheh15:37
mupPR pc-amd64-gadget#10 closed: Add mmx64.efi (MokManager) to support mokutil <Created by tsunghanliu> <https://github.com/snapcore/pc-amd64-gadget/pull/10>15:55
mupPR pc-amd64-gadget#11 closed: Add mmx64.efi (MokManager) to support mokutil <Created by tsunghanliu> <https://github.com/snapcore/pc-amd64-gadget/pull/11>15:55
mupPR pc-amd64-gadget#10 opened: Add mmx64.efi (MokManager) to support mokutil <Created by tsunghanliu> <https://github.com/snapcore/pc-amd64-gadget/pull/10>15:56
mupPR pc-amd64-gadget#11 opened: Add mmx64.efi (MokManager) to support mokutil <Created by tsunghanliu> <https://github.com/snapcore/pc-amd64-gadget/pull/11>15:56
mvojdstrand: oh, let me try again16:05
mvojdstrand: tried again, can you please check?16:05
mupPR snapd#6740 opened: cmd/snap-update-ns: refactor of profile application (3/N) <Created by zyga> <https://github.com/snapcore/snapd/pull/6740>16:22
mupPR snapd#6741 opened: [RFC] osutil: make CommandFromCore honor the snapd snap <â›” Blocked> <Created by mvo5> <https://github.com/snapcore/snapd/pull/6741>16:24
mvopedronis: ^-- this should fix some more of the gaps in snapd vs core16:25
mvopedronis: iirc you wanted to also move osutil.ExecFromCore, please let me know the details and I will expand the PR16:25
mupPR snapd#6732 closed: tests: run livepatch on 18.04 as well <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/6732>16:30
jdstrandmvo: got it16:43
cachioChipaca, hey, when you have a time could you please take a look to #669416:48
mupPR #6694: tests: improve how snaps are cached <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/6694>16:48
=== pstolowski is now known as pstolowski|afk
=== pachulo_ is now known as pachulo
zygaWAT17:10
zyga-----17:10
zyga+ cat17:10
zygaEOM17:10
zyga+ Use release-tools/debian-package-builder to interactively fix build17:10
zyga/bin/bash: line 62: Use: command not found17:10
* zyga fixes17:14
mupPR snapd#6742 opened: tests: fix syntax error in here-doc <Simple 😃> <Created by zyga> <https://github.com/snapcore/snapd/pull/6742>17:45
zygaChipaca: https://github.com/snapcore/snapd/pull/6742 if you can please18:10
mupPR #6742: tests: fix syntax error in here-doc <Simple 😃> <Created by zyga> <https://github.com/snapcore/snapd/pull/6742>18:10
Chipacaouch18:10
Chipacahow did that land?18:10
Chipacaah, debug18:11
Chipacaok18:11
zygaChipaca: fun right18:18
zygathank you18:18
mupPR snapd#6742 closed: tests: fix syntax error in here-doc <Simple 😃> <Created by zyga> <Merged by zyga> <https://github.com/snapcore/snapd/pull/6742>18:34
mupPR snapd#6738 closed: tests: check for /snap/core16/current in core16-provided-by-core <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/6738>18:50
WimpressSnapcraft Live is starting in 10 mins - https://twitter.com/snapcraftio/status/111858745482136780818:51
zygaWimpress: mind if I haunt the comment section? :)18:58
mupPR snapd#6743 opened: cmd/snap, client, daemon, store: create-cohort <Created by chipaca> <https://github.com/snapcore/snapd/pull/6743>19:02
Chipacapedronis: ^19:02
Chipacaand EOD for me19:02
Chipacatomorrow: snapstate cohortitation /o\19:02
pedronismvo: still around?19:17
mvopedronis: yes19:25
pedronismvo: introducing DeviceContext is going well, but I wonder how much you would dislike this change: https://paste.ubuntu.com/p/hzcxF3rCbS/19:30
pedronismvo: needing Model in the bowels of doInstall is a bit annoying/fragile19:30
mvopedronis: looking19:31
mvopedronis: that looks fine19:32
pedronisok, thx19:32
pedronisit simplifies some things and avoid some strange effects19:32
mvopedronis: thanks for this - do you have advice for osutil.CommandFromCore for me? iirc you wanted this to go to cmd.CommandFromSystemSnap or something like this?19:32
mvopedronis: yeah, it sounds nice19:32
pedronisyes, something in cmd19:33
* mvo nods19:34
pedronismvo: close to InternalToolPath basically19:34
mvopedronis: will do19:37
mvopedronis: thank you!19:37
Wimpresszyga: Thanks for joining us!19:42
zygapleasure to do so, really enjoy this part ;)19:42
zyga:)19:42
Wimpress:-D19:43
zygajdstrand: I know you are busy but may I interest you with https://github.com/snapcore/snapd/pull/671419:44
mupPR #6714: cmd/snap-confine: reject crafted /tmp/snap.$SNAP_NAME <Created by zyga> <https://github.com/snapcore/snapd/pull/6714>19:44
mvopedronis: hm, we have an import cycle if I move CommandFromCore to cmd19:44
pedronismvo: hm19:45
mvopedronis: snapinfo.go is the problem19:45
mvopedronis: it imports client which in turn imports snap which imports snapfs19:45
pedronisblargh19:46
mvopedronis: maybe I can move snapinfo out?19:46
mvoit does not fit there much19:46
pedronisyes, the problem is that so far we haven't found where to put it either19:46
pedronisalso we do plan to break snapfs and snap19:46
pedronisbut dep but not now19:46
pedroniss/but dep/dep/19:46
mvopedronis: I could move CommandFromCore into its own cmd/cmdutil pkg?19:47
mvopedronis: together with InternalToolPath maybe19:47
pedronisthat sounds ok, if we move both, at least temporarely, we really need to find a home for snapinfo.go19:48
pedronisbut I haven't had time to think one yet19:48
mvopedronis: ok, I will explore this avenue19:49
zygapedronis: do you think you will have time to look at https://github.com/snapcore/snapd/pull/6717 before your holidays20:25
mupPR #6717: snap: fix interface bindings on implicit hooks <Created by zyga> <https://github.com/snapcore/snapd/pull/6717>20:25
zygapedronis: alternatively, can you defer to mvo for review?20:25
pedroniszyga: it's on my list to try to review tomorrow20:27
zygathank you!20:27
mvopedronis: 6741 is now updated, probably does not need your review I tried to follow your guidance20:30
* mvo -> sleep20:30
mupPR snapd#6744 opened: tests: make test parallel-install-interfaces work for boards with pre-installed snaps <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/6744>21:12
mupPR snapd#6745 opened: tests: make snap-connections test work on boards with snaps pre-installed <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/6745>21:50
mupBug #1825254 opened: auto-complete doesn't work on ubuntu core 18 <Snappy:New> <https://launchpad.net/bugs/1825254>22:21
mupPR snapd#6746 opened: cmd: typedef struct sc_error <Simple 😃> <Created by zyga> <https://github.com/snapcore/snapd/pull/6746>22:44

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!