/srv/irclogs.ubuntu.com/2020/02/12/#snappy.txt

mupPR snapd#8115 closed: overlord/devicestate: fix preseed unit tests on systems not using /snap <Simple 😃> <Skip spread> <Created by bboozzoo> <Merged by anonymouse64> <https://github.com/snapcore/snapd/pull/8115>00:56
mupPR snapcraft#2926 closed: plugin handler: process elf files only if base is specified <Created by cjp256> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/2926>01:32
mupPR snapcraft#2928 closed: logging: use .warning instead of deprecated .warn <Created by sergiusens> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/2928>01:32
=== hggdh is now known as hggdh-msft
mupPR snapcraft#2931 opened: store: improve platform detection <Created by sergiusens> <https://github.com/snapcore/snapcraft/pull/2931>04:11
mborzeckimorning06:29
mvohey mborzecki06:34
mborzeckimvo: hey06:34
mborzeckimvo: looks like despite all the workarounds paplay still gets stuck06:34
mborzeckiquick round trip to schoold, back in 3006:37
mvomborzecki: meh, that's sad06:37
mvomborzecki: see you in 30"06:37
mborzeckire07:09
pstolowskimorning08:00
mborzeckipstolowski: hey08:01
mvohey pstolowski08:06
mvopstolowski: looks like 8102 is super close, just one question in there that smeels like a followup, I'm in favor of merging (also it's green :)08:13
pstolowskimvo: hey. zyga raised a valid concern there, i need to think08:15
zygapstolowski: I think the way the code worked before was more flexible08:48
zygapstolowski: and I fear if you add the right test now we will see it is not working correctly :/08:49
zygapstolowski: perhaps we should keep it like it was before, in ifacestate but based on repo data08:49
zygapstolowski: that is a far safer 2.44 patch08:49
pstolowskizyga: i could use exported GuessCoreSnap() from repo like i had initially in that branch. but i wonder if what we had before had any practical significance since we do restart into new snapd/core?08:51
zygapstolowski: that question requires analysis - I don't know08:51
zygapstolowski: but I do know the old mode reacted instantly while the new mode reacts on restart08:51
* pstolowski quick erran, bb shortly08:56
pstolowski*errand08:56
mborzeckianyone seen centos-7 fail in recent spread runs?09:08
mborzeckihmm looks like it's working now09:10
zygaI didn't09:14
mupPR snapd#8121 opened: spread: move centos to stable systems <Simple 😃> <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/8121>09:20
zygamborzecki: can you review https://github.com/snapcore/snapd/pull/811309:22
mupPR #8113: cmd/snap-confine: bring /var/lib/dhcp from host, if present <Bug> <Created by zyga> <https://github.com/snapcore/snapd/pull/8113>09:22
mborzeckisure09:22
mborzeckizyga: hm should this be done in interfaces/network_control.go as a mount entry?09:25
zygamborzecki: I don't think so, it feels like a standard thing you can get to with devmode09:26
zygabut09:26
zygait's an idea09:26
zygaperhaps?09:26
zygadunno09:26
mborzeckizyga: right, but with devmode you can also access fonts and we still add a mount when desktop is connected09:27
mborzeckiwell, absically so that it ends up in the right place09:27
zygamborzecki: that's a valid point of view09:28
mborzeckizyga: ok, let me add a comment and maybe we can discuss it more09:29
zygaok09:29
mborzeckizyga: if we have add a mount in MountConnectedPlug(), it should behave the same right? however, the iface is not auto connected, but even so, without the connection on AA enabled systems the process won't be able to access /var/lib/dhcp either09:33
zygamborzecki: I think so, there are subtle differences between the two09:43
zygamborzecki: but I think it could work out okay09:43
mvo8110 needs a second review (should be easy)09:50
* zyga looks09:50
zygamvo: reviewed, please look09:53
mvozyga: thank you09:54
* zyga goes to prepare the nvidia patch09:55
mupPR snapd#8063 closed: cmd/snap: implement 'snap remove-user' <Created by chipaca> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8063>10:01
mupPR snapd#8110 closed: store: add support for resume in DownloadStream <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8110>10:02
mborzecki#8060 needs a 2nd review, should be fairly simple10:19
mupPR #8060: gadget: skip update when mounted filesystem content is identical <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/8060>10:19
zygabooo10:19
zygaI already did10:19
zygatwo weeks of waiting10:19
zygaman10:19
mborzeckisame for #804710:19
mupPR #8047: tests: detect LXD launching i386 containers <Test Robustness> <Created by zyga> <https://github.com/snapcore/snapd/pull/8047>10:19
pedronismvo: done10:31
pstolowskire10:38
mborzeckimvo: hm crazy idea to consider if you're feeling brave: https://github.com/snapcore/snapd/pull/8105/files#r37816995610:41
mupPR #8105: store: detect if server does not support http range headers <Created by mvo5> <https://github.com/snapcore/snapd/pull/8105>10:41
mvomborzecki: yeah, love this10:42
mborzeckimvo:  no need to truncating and overwriting what's already there10:42
mupPR snapd#8108 closed: tests/main/interfaces-pulseaudio: use custom pulseaudio script, set kill timeout <Test Robustness> <Created by bboozzoo> <Merged by bboozzoo> <https://github.com/snapcore/snapd/pull/8108>10:48
pedronispstolowski: hi, https://github.com/snapcore/snapd/blob/master/overlord/snapstate/handlers.go#L135110:52
pedronissorry10:52
pedronispstolowski: I meant, https://github.com/snapcore/snapd/pull/8102#discussion_r37817577510:52
mupPR #8102: o/ifacestate: move ResolveDisconnect to ifacestate <Needs Samuele review> <⛔ Blocked> <Created by stolowski> <https://github.com/snapcore/snapd/pull/8102>10:52
mvothanks pstolowski, ijohnson, mborzecki and pedronis for your suggestions in 8119! much appreciated your excellent feedback10:53
* mvo waits for it to become green10:53
zygare10:57
zygasorry, had a small unexpected errand at home (drilling)10:57
zygareproduced nvidia issue10:57
zygaworking on a fix10:57
mborzeckizyga: through drilling?10:58
zygamborzecki: $wife needed to install new baby gate thing (we have few doors in our house so everything is unprotected and open) and the gate arrived now instead of next week10:58
pstolowskipedronis: thank you. i thought we always restart but wasn't sure if it answers the concern, and if that's all there is to it10:59
mvozyga: nice, thank you!11:02
pedronispstolowski: I actually think we might have a bug, but in the other direction, repo thinks we are using snapd too early11:11
pstolowskipedronis: hmm you might be right11:13
mupPR snapd#8122 opened: interfaces/opengl: allow datagrams to nvidia-driver <Bug> <Created by zyga> <https://github.com/snapcore/snapd/pull/8122>11:13
zygamborzecki: ^11:13
pedronispstolowski: because the actuall connections don't get switch over until reloadConnections happens early at start, but the repo will think we are using snapd as soon as we do setup-profiles for it?11:17
mvomborzecki: 7972 is green now, I guess it can be merged?11:19
mborzeckimvo: yes11:21
zygamvo: please consider 8122 for the point release11:22
zygamvo: it's not urgent but feels safe as it is just extra permissions for opengl specific to nvidia11:22
zygamvo: I guess you can consider it pre-reviewed by jamie as he proposed the change in the first place11:22
mupPR snapd#7972 closed: overlord/snapstate, wrappers: undo of snapd on core <Remodel 🚋> <Created by bboozzoo> <Merged by bboozzoo> <https://github.com/snapcore/snapd/pull/7972>11:23
mvozyga: yeah, I think that's sensible11:23
mvopedronis: concerns for 8122 for 2.43.3? looks small and safe11:25
zygamborzecki: https://github.com/snapcore/snapd/pull/8121#issuecomment-58516178211:27
zygaboo11:27
mupPR #8121: spread: move centos to stable systems <Simple 😃> <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/8121>11:27
mborzeckihahah11:27
mborzeckipfff but really, `spread -list google-unstable:` erorrs out right away, so there's no way to tell whetner there are any tests to run even11:30
pstolowskipedronis: sorry, i had to check the code.. yes, this looks problematic11:31
mborzeckizyga: pushed a workaround11:36
mupPR snapd#8116 closed: test/lib/user: add helper lib for doing things for and as a user <Created by bboozzoo> <Merged by bboozzoo> <https://github.com/snapcore/snapd/pull/8116>11:40
zygamborzecki: I've reworked the dhcp PR to do what you suggested, altered appropriate tests11:42
zygafingers crossed :)11:42
zygaI'll grab some tea11:42
mborzeckizyga: does it work still? :)11:42
zygamborzecki: should, I'll know in 30 minutes11:43
zygabrb11:43
mborzeckihaha11:43
mborzeckibtw. maybe i should post my cloud-init configs for prebuilding spread images11:43
ogramwhudson, are you still the go-to guy for subiquity bugs ?11:48
zygamborzecki: do!11:53
pedronispstolowski: not with urgency but we need to slowly try to reduce the need for guess* in repo and/or pass the value in from ifacestate12:03
zygamborzecki: it's a little harder than the other one but I should have an alternate PR shortly12:06
pstolowskipedronis: ack, will do soon. going to land resolve-disconnect PR then (after addressing remaining comments)12:10
mupPR snapcraft#2931 closed: store: improve platform detection <Created by sergiusens> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/2931>12:13
zygamborzecki: https://github.com/snapcore/snapd/pull/812312:28
mupPR #8123: interfaces/network-control: bring /var/lib/dhcp from host <Bug> <Created by zyga> <https://github.com/snapcore/snapd/pull/8123>12:28
mupPR snapd#8123 opened: interfaces/network-control: bring /var/lib/dhcp from host <Bug> <Created by zyga> <https://github.com/snapcore/snapd/pull/8123>12:28
zygamborzecki: some tests are still running locally but that's the rough idea12:29
mborzeckizyga: thanks, will take a look12:33
zygahmm12:36
zygadoesn't work ..12:36
zyga(regression test failed)12:36
* zyga looks12:36
zygaah, I didn't connect the interface in the test12:40
zygamborzecki: it's weird that this passed in the other one12:40
zygaare all tests running with --devmode?12:40
zygaone more try12:53
zygamborzecki: oh boy13:14
zygaupdate.go:85: cannot change mount namespace according to change mount (/var/lib/snapd/hostfs/var/lib/dhcp /var/lib/dhcp none rw,bind 0 0): cannot write to "/var/lib/snapd/hostfs/var/lib/dhcp" because it would affect the host in "/var/lib/snapd"13:14
pedronispstolowski: I did a pass on #804613:17
zygagot lots of - Fetch and check assertions for snap "core" (8683) (cannot get device session from store: store server returned status 400 and body "{\"error_list\":[{\"code\":null,\"message\":\"Nonce is missing or invalid.\"}],\"errors\":[\"Nonce is missing or invalid.\"],\"result\":\"error\"}\n")13:17
mupPR #8046: many, tests: integrate all preseed bits and add spread tests <Complex> <Needs Samuele review> <Preseeding 🍞> <Created by stolowski> <https://github.com/snapcore/snapd/pull/8046>13:17
pstolowskipedronis: ty13:18
pedroniszyga: store was redeploying something, maybe it's related13:19
mupPR snapd#8105 closed: store: detect if server does not support http range headers <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8105>13:19
zygamhm13:21
pedronispstolowski: #8102 can be merged now?13:33
mupPR #8102: o/ifacestate: move ResolveDisconnect to ifacestate <Needs Samuele review> <Created by stolowski> <https://github.com/snapcore/snapd/pull/8102>13:33
pstolowskipedronis: i wanted to address your remark about SystemSnapName never returning ""; but perhaps will do in a followup13:35
mupPR snapd#8124 opened: tests: Fix core revert channel after 2.43 has been released to stable <Skip spread> <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/8124>13:35
pstolowskipedronis: ok if i squash-merge #8102 in case of something unexpected and a need for revert?13:41
mupPR #8102: o/ifacestate: move ResolveDisconnect to ifacestate <Needs Samuele review> <Created by stolowski> <https://github.com/snapcore/snapd/pull/8102>13:41
zygamborzecki: please check https://github.com/snapcore/snapd/pull/812313:44
mupPR #8123: interfaces/network-control: bring /var/lib/dhcp from host <Bug> <Created by zyga> <https://github.com/snapcore/snapd/pull/8123>13:44
zygamborzecki: it's an interesting development13:44
zygamborzecki: if accepted we could remove more things from snap-confine13:44
zygamborzecki: and move them to the correct interface13:44
mupPR snapcraft#2932 opened: elf: resolve paths in `ldd()` to purge relative path components <Created by cjp256> <https://github.com/snapcore/snapcraft/pull/2932>13:46
zygamborzecki: https://github.com/snapcore/snapd/pull/8121 is still red13:48
mupPR #8121: spread: move centos to stable systems <Simple 😃> <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/8121>13:49
zygamborzecki: I'd like to look at https://github.com/snapcore/snapd/pull/812113:49
zygamborzecki: it produces some selinux denials13:49
zygamborzecki: I'll try to fix them now but I may need your eyes later13:50
mupPR pc-amd64-gadget#34 opened: Makefile: add "regex" to the modules for the non-uefi grub <Created by mvo5> <https://github.com/snapcore/pc-amd64-gadget/pull/34>13:50
ijohnsonmorning folks13:51
zygagood morning13:51
ijohnsonhey zyga13:51
mborzeckiijohnson: hey13:51
mupPR snapd#8119 closed: httputil: add NoNetwork(err) helper, spread test and use in serial acquire <Squash-merge> <⚠ Critical> <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8119>13:51
mupPR snapd#8122 closed: interfaces/opengl: allow datagrams to nvidia-driver <Bug> <Created by zyga> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8122>13:57
zygathank you mvo!13:58
* cachio afk14:09
zygamvo, xnox: do you know what triggers cgroup v2 mode in systemd? will we auto-transition to v2 at some point or is that an explicit toggle?14:13
zygaI'm trying to understand if ubuntu core 20 will be cg-v1 or cg-v214:13
xnoxzyga:  ubuntu 20.04 LTS will be hybrid14:14
xnoxfedora latest is v2-only14:14
xnoxhybrid is the same as what bionic is, i believe14:14
zygaright, but is that configuration or some kind of auto-detection in systemd14:14
zygabased on something (kernel version?)14:14
xnoxno14:14
xnoxthis is a build-time systemd option14:15
zygaah, I see14:15
zygathank you!14:15
xnoxhowever, a user can override it with a kernel cmdline option14:15
zygado you thin core 22 will be v2?14:15
zyga*think14:15
xnoxso for example on focal you can test and boot in v2-only mode if you want to prepare for the future / fedora14:15
xnoxi think fedora/rhel/suse will be v2-only before core22 ships14:15
mborzeckizyga: systemd.unified_cgroup_hierarchy=114:16
zygamborzecki: I do know this :)14:16
zygaI was wondering what kind of future is ahead14:16
mborzeckixnox: do you think rhel8 will become v2 at some point, or rather rhel9?14:17
mupPR pc-amd64-gadget#34 closed: Makefile: add "regex" to the modules for the non-uefi grub <Created by mvo5> <Merged by xnox> <https://github.com/snapcore/pc-amd64-gadget/pull/34>14:18
mborzeckizyga: do you have any graphical snaps on TW? like gnome-logs, gnome-calculator?14:46
mborzeckizyga: if so, do the fonts render correctly?14:46
zygayes14:47
zygalet me check14:47
mupPR snapd#8125 opened: data/selinux: unify tabs/spaces <Simple 😃> <Created by zyga> <https://github.com/snapcore/snapd/pull/8125>14:47
zygamborzecki: ^ can you look at that quickly please14:47
zygabooting tw14:48
zygamborzecki: gnome-calcuator runs ok14:50
zygawell14:50
zygaalmost14:50
zygathe x in the corder is broken14:50
zyga(close window)14:50
zygagnome-logs also works and has the same issue14:51
mborzeckizyga: hm i still get boxes instead of fonts14:56
=== alan_g is now known as alan_g_
zygaback with tea15:14
zygamborzecki: hm15:14
zygaI should install arch and try15:15
zygamborzecki: did you you try to debug it?15:15
pstolowskioh well travis is super slow15:31
mvoyeah, it's terrible when trying to make a release15:34
* zyga goes for lunch/dinner 15:35
zygaback16:30
zygafunny day - everyone had something else for dinner16:30
zygamvo: any luck with the release?16:31
mupPR snapd#8126 opened: release: 2.43.3 <Created by mvo5> <https://github.com/snapcore/snapd/pull/8126>16:41
mvozyga: it's almost out16:43
zygaI see the tarball16:44
zygaI'll make a suse package16:44
mvoI need to create the proper github release but in a meeting right now so a bit slow16:45
zygaah wait, I didn't read the github page properly16:45
zygasure16:45
zygaI'll wait16:45
mvozyga: should be ready in ~15min16:45
mvozyga: please have a look16:48
zygalooking16:48
zygayep16:48
zygamaking the package now16:48
zygathank you :)16:48
zygabuilding16:51
=== alan_g_ is now known as alan_g
zygasuse updated17:02
zygathere's a small patch back to master17:02
zygaI'll open it shortly17:02
zygajdstrand: can you try to look at https://github.com/snapcore/snapd/pull/8113 and https://github.com/snapcore/snapd/pull/8123 (they form a either-or pair)17:14
mupPR #8113: cmd/snap-confine: bring /var/lib/dhcp from host, if present (approach a) <Bug> <Created by zyga> <https://github.com/snapcore/snapd/pull/8113>17:15
mupPR #8123: interfaces/network-control: bring /var/lib/dhcp from host (approach b) <Bug> <Created by zyga> <https://github.com/snapcore/snapd/pull/8123>17:15
zygahuh17:16
zygaFeb 12 15:50:58 feb121515-669259 systemd[5577]: snapd.service: Failed to execute command: Exec format error17:16
zygaFeb 12 15:50:58 feb121515-669259 systemd[5577]: snapd.service: Failed at step EXEC spawning /snap/snapd/x2/usr/lib/snapd/snapd: Exec format error17:16
zygamvo: ^17:16
zygaI'm seeing this in CI17:16
mvozyga: uh, related to the release?17:27
mvocachio: 2.43.3 snapd snap is in beta now, 2.43.3 core should be available soon17:35
cachiomvo, nice, I'll start right now17:35
cachiothnaks17:35
mvocachio: thank you!17:35
mupPR snapcraft#2933 opened: remote-build: introduce --launchpad-snapcraft-channel option <Created by cjp256> <https://github.com/snapcore/snapcraft/pull/2933>17:37
pedronismvo: did you backport the minor nvidia version fix?18:02
mvopedronis: it's not in18:08
mvopedronis: which is slightly silly, I was too cautious before and then forgot18:09
pedronisit's ok I suppose, but checked because I was confused when looking at the changelog18:09
mvopedronis: yeah, we talked about adding it18:10
mvopedronis: and I was unsure (a bit) and did not cherry pick right away and then it felt through the cracks18:10
mupPR snapd#8121 closed: spread: move centos to stable systems <Simple 😃> <Created by bboozzoo> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8121>18:12
ograheh, funny ... "snap find 'screen recorder'" finds xonotic ... i wonder why18:15
techalchemyzyga, thanks for the fix!18:19
jdstrandzyga: those aren't showing up in https://github.com/snapcore/snapd/pulls/review-requested/@me (for me :)18:38
jdstrandzyga: please add the 2.44 milestone if you want them there unless you are seeing these are emergency18:38
cachioijohnson, hey18:49
cachioI updated the spread pr18:50
cachiothe problem with the tests is that there is not unit tests at all for google backend18:50
cachioijohnson, for testing what I did is to use the -vv command for spread and iterate over differnt images to see if the selected is the desired one18:51
cachioijohnson, the problem is that the images change over time18:51
cachioso it is difficult to test that in a spread test d18:51
cachiozyga, when you have time could you please take a look to https://github.com/snapcore/snapd/pull/790020:18
mupPR #7900: tests: do reset of tests during restore and add checks to validate the fs <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/7900>20:18
cachioI remember you developed a tool20:18
zygajdstrand: ack, not an emergency, just curious choice20:28
zygatechalchemy: pleasure :-)20:28
zygacachio: sure, I’ll look in an hour20:29
cachiozyga, thanks20:29
zygacachio: looks very interesting, I will review it in detail first thing tomorrow21:09
zygaToday is too late for a detailed review21:09
zygaI have some comments but I am happy to see progress on robustness21:09
cachiozyga, nice, thanks21:11
cachioit needs some improvements but I think it could help21:11
cachiozyga, see you tomorrow21:11

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