/srv/irclogs.ubuntu.com/2019/08/20/#snappy.txt

mborzeckimorning05:01
mborzeckimvo: hey05:54
mvomborzecki: hey, good morning05:56
mborzeckimvo: found a bug in snap info, try `snap install hello-world hello-world_foo`, then compare `snap info hello-world` and `snap info hello-world_foo`05:56
mborzeckimvo: the fix is super simple so no big deal06:01
mvonice06:01
mborzeckimvo: also this request sounds sane https://forum.snapcraft.io/t/feature-request-link-to-store-page-in-snap-info/1283806:02
mvoindeed06:04
mborzeckii'll open PRs for both in a bit06:07
mvota06:08
mupPR snapd#7274 closed: tests: change cgroups so that LXD doesn't have to <Created by zyga> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/7274>06:16
mupPR snapd#7257 closed: packaging: fix symlink for snapd.session-agent.socket <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/7257>06:38
mborzeckisurprisingly the hardest part of the fix is figuring out the whitspace in channel output and disabling whitespace-cleanup :/06:51
mupPR snapd#7275 opened: cmd/snap: fix remote snap info for parallel installed snaps <Simple πŸ˜ƒ> <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/7275>06:53
mborzeckimvo: ^^06:54
mvomborzecki: thanks, looking06:56
mupPR snapd#7124 closed: many: create system-usernames user/group if both don't exist <Created by jdstrand> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/7124>06:56
mvoI just branched 2.41 locally07:01
=== pstolowski|afk is now known as pstolowski
pstolowskimorning07:03
mborzeckipstolowski: hey07:03
mvohey pstolowski07:13
zygagood morning07:19
zygasorry for being late, Lucy is very active today and I had to help more07:19
pstolowskio/07:19
zygahey :)07:19
* zyga runs TW updates07:20
mvohey zyga07:20
zygagood morning, how are you doing todayt?07:20
mvozyga: good, a bit sleepy but good, today is 2.41~pre1 day!07:21
zygaexcellent, time to roll those releases :)07:21
mborzeckizyga: hey07:36
zyga:)07:36
* zyga fights gnome-shell07:36
mborzeckizyga: the year of linux desktop07:37
zygaI can either get 100% scaling and background or 200% scaling and solid white background07:37
zygain x mode, in wayland the resolution is stuck but I have background07:37
mborzeckizyga: ui scaling?07:37
zygayeah07:37
mborzeckiheh07:38
zygahey pedronis, good morning07:39
mborzeckizyga: fwiw, who needs to look at the background ;) all my windows are full screen07:39
zygamborzecki: white BG at 500 nits can hurt07:39
mborzeckizyga: it's called front camera flash in smartphones :)07:40
mborzeckipedronis: hi, opened the PR with renames yday, https://github.com/snapcore/snapd/pull/7273 in case you want to look, or i can poke others for reviews07:41
mupPR #7273: gadget, overlord/devicestate: rename Position/Layout <Gadget update> <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/7273>07:41
pedronismborzecki: I'll do a quick scan07:43
mborzeckipedronis: thanks!07:44
=== pedronis_ is now known as pedronis
pedronismborzecki: made a couple of remarks, about fields07:49
pedronislet me know if they make sense07:49
pedronisor not07:50
mborzeckipedronis: s/StructureLayout/Layout/ sounds fine, but s/ContentLayout/Content/ is tricky, VoluemContent already has a field named Content, maybe RawContentLayout/RawLayout would work?07:55
pedronismborzecki: why is tricky?07:56
pedronis(I'm missing something here)07:56
pedronismborzecki: what is your worry?07:59
mborzeckipedronis: LaidOutStructure embeds *VolumeStructure which alreay has a field named Content (that describes all content, raw and filesystem), thus suggestion to use Raw(Content,Layout) which is only present for raw content anyway07:59
pedronismborzecki: no, then the boring LaidOutContent is probably my preference08:01
mborzeckipedronis: haha, boring works for me too :)08:02
mupPR snapd#7276 opened: cmd/snap: include snapcraft.io page URL in snap info output <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/7276>08:11
mborzeckizyga: https://lists.opensuse.org/opensuse-project/2019-08/msg00011.html08:12
zygayeah, I saw that earlier08:12
zygabrb08:17
zygageez, my son is still sleeping08:20
mborzeckiChipaca: thanks for the reviews, is default store check only possible within snapd?08:41
Chipacamborzecki: needs a device auth context i think08:42
Chipacamborzecki: meaning, yeah08:46
mborzeckiChipaca: ok, then the store-url could be added to the snapd api on /v2/snaps/<snap>08:46
Chipacamborzecki: yes08:46
Chipacamborzecki: would be even nicer if it came from remote08:46
* zyga quick breakfast08:47
mborzeckiChipaca: heh ;)08:49
mupPR snapd#7275 closed: cmd/snap: fix remote snap info for parallel installed snaps <Simple πŸ˜ƒ> <Created by bboozzoo> <Merged by bboozzoo> <https://github.com/snapcore/snapd/pull/7275>08:50
zygare08:51
Chipacamborzecki: #1840751 FWIW08:59
mupBug #1840751: Please include store-url when it exists <Snap Store:New> <https://launchpad.net/bugs/1840751>09:00
mborzeckiChipaca: thank you!09:00
zygayay, I think I'm down to 0 leaks on all classic systems09:17
zygacore seems to leak something odd09:17
zyganot snaps visible in snap list09:17
zygabut actual mounted squashes and loopback devices09:17
zygaI have three cleanup patches upcoming, just let the classic run finish again09:17
zygaand looking at core, core tests don't overlap with classic fully so I bet there are some I just didn't look at09:17
pedronismvo: I hope I fixed 7135, not quite sure what happened with that assertion09:22
mupPR snapcraft#2671 opened: docker: remove snapcraft-wrapper <Created by abitrolly> <https://github.com/snapcore/snapcraft/pull/2671>09:30
mvopedronis: thanks09:32
mupPR snapcraft#2672 opened: docker: use apt-get to avoid warnings <Created by abitrolly> <https://github.com/snapcore/snapcraft/pull/2672>09:39
zygare09:41
mupPR snapd#7277 opened: [RFC] overlord/snapstate: fix undo on firstboot seeding <Created by stolowski> <https://github.com/snapcore/snapd/pull/7277>09:42
pstolowskipedronis: ^ a high-level ack/nack on the apporach would be great09:44
pstolowski*approach09:45
pstolowskizyga: sorry for pestering... #7081 waves at you ;)09:49
mupPR #7081: ifacestate: optimize auto-connect by setting profiles once after all connects <Needs Samuele review> <Created by stolowski> <https://github.com/snapcore/snapd/pull/7081>09:49
zygaindeed, after this call I'll jump right in09:49
mupPR snapcraft#2673 opened: Single dockerfile <Created by abitrolly> <https://github.com/snapcore/snapcraft/pull/2673>09:51
pedronispstolowski: btw, there are still a couple of things to do here: https://trello.com/c/yxC3omC9/245-snap-snapctl-unset-support10:05
pstolowskipedronis: yes, i saw your comment, thanks, i'm on it10:05
mborzeckiChipaca: https://paste.ubuntu.com/p/sFtP2kQ5M6/ not sure about the part that finds whether it's a custom store or not10:17
pedronismborzecki: I think we shouldn't do anything until we have store support10:23
mupPR snapd#7031 closed: debian: re-enable systemd environment generator <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/7031>10:26
mborzeckipedronis: yeah, you're probably right, i'll keep the branch so that the client bits are ready when store updates the api10:28
mupPR snapd#7278 opened: tests: unmount fusectl after testing <Simple πŸ˜ƒ> <Created by zyga> <https://github.com/snapcore/snapd/pull/7278>10:28
mupPR snapd#7279 opened: tests: restore nsdelegate clobbered by LXD <Simple πŸ˜ƒ> <Created by zyga> <https://github.com/snapcore/snapd/pull/7279>10:29
mupPR snapd#7276 closed: cmd/snap: include snapcraft.io page URL in snap info output <Created by bboozzoo> <Closed by bboozzoo> <https://github.com/snapcore/snapd/pull/7276>10:30
mupPR snapd#7264 closed: packaging/fedora: build on RHEL8 (2.40) <Simple πŸ˜ƒ> <Created by bboozzoo> <Merged by bboozzoo> <https://github.com/snapcore/snapd/pull/7264>10:47
mupPR snapd#7280 opened:  debian: re-enable systemd environment generator (2.41) <Created by mvo5> <https://github.com/snapcore/snapd/pull/7280>10:52
* zyga coffee10:55
Chipacajdstrand, zyga, mvo, fun coffee read: https://lwn.net/Articles/796687/11:05
zygaChipaca: I take that and raise https://lwn.net/Articles/796641/11:07
mupPR snapd#7281 opened: cmd/snap: fix snap unset help string <Simple πŸ˜ƒ> <Created by stolowski> <https://github.com/snapcore/snapd/pull/7281>11:09
mvoChipaca: nice, that was on my reading list11:09
Chipacazyga: fuse tangled into the package manager11:09
mvoChipaca: nice to hear that its apparently worth it :)11:09
Chipacazyga: what could go wrong?11:10
Chipacamvo: :)11:10
zygaChipaca: squashfs :)11:10
Chipacazyga: fuse pinning a core, but everywhere all the time11:11
Chipacaand that's when it's working :)11:11
Chipacaanyway, will read more11:11
Chipacamvo: when was core 4486 from?11:15
Chipacanever mind, over a year old11:16
* Chipaca used 'git log' to get a good enough guess11:16
Chipacaok i'm off to look lunch in the face11:25
mupPR snapd#7247 closed: interfaces/mount: forward SNAPD_DEBUG to ns tools <Simple πŸ˜ƒ> <Created by zyga> <Closed by zyga> <https://github.com/snapcore/snapd/pull/7247>12:51
mupPR snapd#7282 opened: asserts: move Model to its own model.go <Simple πŸ˜ƒ> <Created by pedronis> <https://github.com/snapcore/snapd/pull/7282>12:51
diddledanseems futexes are blocked but silently... https://www.irccloud.com/pastebin/rolINucz/12:59
diddledanI can't find any mention of seccomp blocking the syscall in /var/log12:59
=== ricab is now known as ricab|lunch
mupPR snapcraft#2664 closed: cli: handle exception when cleaning a part with a fresh project <Created by cjp256> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/2664>13:04
mupPR snapcraft#2666 closed: tests: move meta testing to its own package <Created by sergiusens> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/2666>13:04
mupPR snapcraft#2667 closed: yaml utils: move OctInt from meta <Created by sergiusens> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/2667>13:04
Chipacamvo: https://irclogs.ubuntu.com/2019/08/19/%23snappy.html#t15:0813:06
mupPR snapd#7283 opened: tests: unmount binfmt_misc on cleanup <Simple πŸ˜ƒ> <Created by zyga> <https://github.com/snapcore/snapd/pull/7283>13:19
mupPR snapd#7278 closed: tests: unmount fusectl after testing <Simple πŸ˜ƒ> <Created by zyga> <Merged by zyga> <https://github.com/snapcore/snapd/pull/7278>13:26
zygamvo: is this what you mentioned a moment ago?13:41
zygahttps://www.irccloud.com/pastebin/e89eTfcR/13:41
Chipacazyga: what was this about snap info of files?13:47
Chipacasomething to do with The Pope(y)13:47
pedronisChipaca: I think we might check if a dir exist before doing anything else, that does unexpected/expected things if you happen to be sitting somewhere with a dir foo which contains snap foo13:48
Chipacahmm, i thought i'd checked for that, but … maybe? is there a concrete bad case?13:49
pedronisChipaca: it's not a priority either way, please queue it13:50
Chipacathere is some weird behaviour if you have a directory foo and do 'snap info foo' and foo is an installed snap that the store knows about and the foo directory does not contain a snap13:50
Chipacae.g., snap install http go; cd /snap/go/current/src/net; snap info http β†’ no channel map; weird13:51
jdstrandChipaca: re lwn> on the agenda for lunch :)13:51
Chipacajdstrand: :)13:51
mupPR snapcraft#2659 closed: Add support for building inside podman containers <Created by abitrolly> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/2659>13:52
Chipacabut, i'd like to know if that is the strangeness or if there's something worse13:52
Chipacapopey: ^13:52
mvozyga: yes13:52
diddledanjdstrand: I presume futex is being blocked somewhere but I can't find any log: https://www.irccloud.com/pastebin/rolINucz/14:15
mupPR snapd#7284 opened: tests: clean user and group for test system-usernames-install-twice <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/7284>14:22
=== ricab|lunch is now known as ricab
jdstranddiddledan: the futex syscall is allowed by default. the man page for futex has some things to say about EPERM. specifically: "(FUTEX_UNLOCK_PI)  The  caller  does not own the lock represented by the futex word" and "(FUTEX_LOCK_PI, FUTEX_TRYLOCK_PI, FUTEX_CMP_REQUEUE_PI)  The  caller is  not  allowed  to  attach  itself  to the futex at uaddr". you may want to see the man page for more context14:23
diddledanhmm14:23
jdstrandfor the last point, it also says "(This may be caused by  a state corruption in user space.)14:24
diddledanI wonder what mycroft is doing that it is failing14:24
jdstrand"14:24
diddledansomething in the voice/speech part of mycroft is reporting this https://www.irccloud.com/pastebin/crszJzeL/14:25
diddledanso in response I did an strace which only shows futex failing14:25
jdstrandhmm14:27
diddledanI'm not sure where that error lies in the mycroft code though, because it's huge :-)14:27
diddledanmycroft itself is python but it has a few C-like libraries that it calls - I did a bit of research and found references to C++14:28
diddledanresearch on the specific error message, I mean14:29
mupPR snapd#7285 opened: httputil: improve http2 PROTOCOL_ERROR detection <Created by mvo5> <https://github.com/snapcore/snapd/pull/7285>14:29
jdstranddiddledan: does it use the preload for the semaphore? it is python multithreading that needs that.... I would've expected a denial. it is possible the kernel is rate limiting it. you could: sudo apparmor_parser -r /var/lib/snapd/apparmor/profiles/snap.mycroft.* ; sudo journalctl --output=short --follow --all | sudo snappy-debug14:31
diddledanI do launch it with the snapcraft-preload, yes14:31
jdstranddiddledan: then try again. the reload will cause denials that are being rate limited to show (until they are rate limited again) and the snappy-debug turns off general kernel rate limiting14:32
jdstranddiddledan: the above might make a denial pop out that was rate limited, so still worth a try14:32
zygaChipaca: afk14:33
Chipacazyga: ?14:34
ijohnsonthanks Chipaca!14:35
ijohnsonChipaca: also do you think the unit tests I have there are sufficient or do I need to add more, etc. ?14:36
mupPR snapd#7286 opened: tests: use images:ubuntu/ in lxd tests <Created by mvo5> <https://github.com/snapcore/snapd/pull/7286>14:36
ijohnson(in #7149 that is)14:37
mupPR #7149: cmd: add snap model command; daemon: add /v2/model, /v2/model/serial REST APIs <Remodel :train:> <Created by anonymouse64> <https://github.com/snapcore/snapd/pull/7149>14:37
Chipacaijohnson: hmm, i didn't look at coverage14:44
Chipacaijohnson: will do in a bit14:44
ijohnsonokay, sounds good14:44
mupPR snapd#7287 opened: hookstate/ctlcmd: snap unset command <Created by stolowski> <https://github.com/snapcore/snapd/pull/7287>14:45
Chipacahmm14:47
mvozyga: fwiw, I think I know why the system-users test is failing, fixing now14:50
zygaOh? I’m eager to know14:57
Chipacaijohnson: maybe i'm not spotting it: what would cause jsonResult to be true?15:01
Chipacaijohnson: in model and serial queries i mean15:02
ijohnsonspecifying ?json=true in a REST API call (i.e. not using the CLI)15:02
mvozyga: PR will be up in some minutes but its boring :)15:02
Chipacaijohnson: thought so. Commented, wrt coverage and other bits and bobs15:03
ijohnsongot it, thanks15:03
Chipacaijohnson: I'd accept these to be in a followup fwiw, if it makes this easier to land15:03
Chipacait's already quite big15:04
ijohnsonI dunno, what does mvo think?15:04
zygare15:04
zygaChipaca: sorry, I noticed the message and replied from my phone15:04
zygaChipaca, popey: you guys should talk about snap info :)15:04
mvoijohnson: I miss some context I think but followup sounds good especially if the pr is already big15:05
mvo2.41 is branched so having it in a not 100% state for a tiny bit is not too terrible15:05
ijohnsonmvo: it's PR #7149, not sure if you wanted to make a review pass on that15:06
mupPR #7149: cmd: add snap model command; daemon: add /v2/model, /v2/model/serial REST APIs <Remodel :train:> <Created by anonymouse64> <https://github.com/snapcore/snapd/pull/7149>15:06
ijohnsonIIRC, this wasn't needed for 2.4115:06
mvoijohnson: yeah, its fine for .4215:07
ogra.42 is the answer !15:07
diddledanis .41 out yet? *ducks the swinging punches*15:09
ijohnsonmvo: so do you want to review that PR as well before landing for 2.42? or am I good to merge once tests are green?15:09
Chipacaijohnson: needs two approvals to merge tho15:10
ijohnsonah I guess Graham only reviewed the help message, I forgot that15:10
mvoijohnson: yeah, what john said, needs two +1 :)15:15
mupPR snapd#7288 opened: tests: cleanup "snap_daemon" user in system-usernames-install-twice <Created by mvo5> <https://github.com/snapcore/snapd/pull/7288>15:27
zygamvo: did you see https://github.com/snapcore/snapd/pull/728415:29
mupPR #7284: tests: clean user and group for test system-usernames-install-twice <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/7284>15:29
mvozyga: meh, I did not15:29
zygayour PR looks more comprehensive :)15:30
zygabut I didn't look deeper yet15:30
mvozyga: yeah, the extrausers bit15:30
mupPR snapcraft#2669 closed: Plugin catkin: forward parallel build count <Created by artivis> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/2669>15:31
* cachio lΓ±unch15:36
mupPR snapd#7281 closed: cmd/snap: fix snap unset help string <Simple πŸ˜ƒ> <Created by stolowski> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/7281>15:42
mupPR snapd#7280 closed:  debian: re-enable systemd environment generator (2.41) <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/7280>15:43
* zyga takes a break15:50
zygaI added service leak detector15:50
zygachecking for stuff that affects systemd serrvices15:51
zyga*services15:51
zygaso looking forward to what the outcome will be15:51
zygaI'll grab coffee and be back to check15:51
=== pstolowski is now known as pstolowski|afk
cachiopstolowski|afk, #7110 updated16:31
cachiothanks16:31
mupPR #7110: tests: new test to check the output after refreshing/reverting core <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/7110>16:31
mupPR snapd#7289 opened: tests: ubuntu 18.10 removed from the google-sru backend on the spread.yaml <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/7289>16:35
zygare16:49
mupPR snapd#7279 closed: tests: restore nsdelegate clobbered by LXD <Simple πŸ˜ƒ> <Created by zyga> <Merged by zyga> <https://github.com/snapcore/snapd/pull/7279>16:50
mupPR snapd#7290 opened: tests: allow test user XDG_RUNTIME_DIR to phase out <Created by zyga> <https://github.com/snapcore/snapd/pull/7290>18:59
* cachio afk19:03
mupPR snapd#7283 closed: tests: unmount binfmt_misc on cleanup <Simple πŸ˜ƒ> <Created by zyga> <Merged by zyga> <https://github.com/snapcore/snapd/pull/7283>19:07
popeyZyga Chipaca can we talk about snap info over mail or a bug? Been on a plane for 11 hours19:27
diddledanwhere you at this week, popey ?19:30
popeyArizona19:30
diddledan\o/19:30
popeyIt am hot19:30
diddledanboo19:31
diddledanI saw your post earlier about 40degrees19:31
mupPR snapd#7291 opened: Add new cases into arch_test <Created by ardaguclu> <https://github.com/snapcore/snapd/pull/7291>19:43
mupPR snapd#7292 opened: interfaces: k8s worker node updates <Created by jdstrand> <https://github.com/snapcore/snapd/pull/7292>23:07
mupPR snapd#7293 opened: interfaces: k8s worker node updates - 2.41 <Created by jdstrand> <https://github.com/snapcore/snapd/pull/7293>23:11

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