[Bug 1851233] Re: building a snap from the eoan tree using the raspi2 defconfig results in gigantic snap package

2019-11-04 Thread Oliver Grawert via ubuntu-bugs
i also tried using the "source-branch: raspi2" argument in the snapcraft.yaml instead of using teh master branch and the "source-tag:" option, but this results in the same big size ... additionally i'd like to remark that due to the transactional nature of UbuntuCore we need to keep the kernel

[Bug 1851233] Re: building a snap from the eoan tree using the raspi2 defconfig results in gigantic snap package

2019-11-04 Thread Oliver Grawert via ubuntu-bugs
yup, seems this is (surprisingly) on by default ... $ grep DEBUG_INFO parts/kernel/build/.config CONFIG_DEBUG_INFO=y # CONFIG_DEBUG_INFO_REDUCED is not set # CONFIG_DEBUG_INFO_SPLIT is not set CONFIG_DEBUG_INFO_DWARF4=y # CONFIG_DEBUG_INFO_BTF is not set i'll try a re-build and override that

[Bug 1851233] Re: building a snap from the eoan tree using the raspi2 defconfig results in gigantic snap package

2019-11-04 Thread Oliver Grawert via ubuntu-bugs
ok, this looks a bit more reasonable now ... $ ls -lh pi4-kernel_raspi2-5.3.0-1011.12_armhf.snap -rw-r--r-- 1 ogra ogra 211M Nov 4 17:05 pi4-kernel_raspi2-5.3.0-1011.12_armhf.snap so CONFIG_DEBUG_INFO=y should better be turned off by default for people building snaps from our trees ;) --

[Bug 1560942] Re: Support swap

2019-11-02 Thread Oliver Grawert via ubuntu-bugs
it is working via manual intervention after login like: https://forum.snapcraft.io/t/enabling-swap-on-ubuntu-core/5440 i.e. i implemented the OS side of it ... but we still do not have any "snap set system.swap=" option to steer it or set via a gadget configuration option in the yaml, this

[Bug 1851233] Re: building a snap from the eoan tree using the raspi2 defconfig results in gigantic snap package

2019-11-06 Thread Oliver Grawert via ubuntu-bugs
the included snapcraft.yaml means i need to have the whole tree locally on disk, as someone who maintains a ton of community images for UbuntuCore i prefer to just have the snapcraft.yaml locally so the disk is only cluttered during build. along with that i am trying to build the eoan tree for the

[Bug 1851233] Re: building a snap from the eoan tree using the raspi2 defconfig results in gigantic snap package

2019-11-06 Thread Oliver Grawert via ubuntu-bugs
well, as a user/customer building my own kernel i'd kind of expect the defconfig for a flavour to not be a debug config. but indeed you are correct, i should have taken a look at the included snapcraft.yaml which, despite being for the wrong flavour (generic/pc- kernel), indeed contains

[Bug 1850977] Re: Snap installs software without user having sudo access

2019-11-14 Thread Oliver Grawert via ubuntu-bugs
policyKit does not involve sudo in any way, it uses systemd-logind from the session to elevate privileges. if you are marked as admin in the policyKit setup you will indeed be able to do admin things no matter what is written in sudoers ;) -- You received this bug notification because you are a

[Bug 1850977] Re: Snap installs software without user having sudo access

2019-11-14 Thread Oliver Grawert via ubuntu-bugs
the requirement for policykit (and dropping of gksu/gksudo) came with the switch to gnome upstream, its a hard requirement for the desktop nowadays. while the default here might be wrong (and should be reviewed by someone from the desktop team), this is definitely not a snapd related bug. i added

[Bug 1852720] Re: docker snap suddenly installed by itself

2019-11-15 Thread Oliver Grawert via ubuntu-bugs
sadly snapd wipes the logs after a while so your snap changes start only at change 33 and this is only a refresh of the already installed docker snap ... it would be interesting to know why/how it got installed in the first place, perhaps someone from the snapd team has a hint here ... do you

[Bug 1852720] Re: docker snap suddenly installed by itself

2019-11-15 Thread Oliver Grawert via ubuntu-bugs
unattended-updates do not deal with snap packages (effectively snaps care for their upgrades themselves, there is no external entity like unattended-upgrades needed)... also snap packages do not "just install themselves", something must have actively triggered that ... along with syslog (or

[Bug 1851233] [NEW] building a snap from the eoan tree using the raspi2 defconfig results in gigantic snap package

2019-11-04 Thread Oliver Grawert via ubuntu-bugs
Public bug reported: typically a kernel snap built from the kernel.ubuntu.com tree using the raspi2 defconfig results in a snap package that is below 150MB in size: $ ls -lh pi2-kernel_raspi2-4.4.0-1110.snap -rw--- 1 root root 129M Jun 4 15:01 pi2-kernel_raspi2-4.4.0-1110.snap trying to

[Bug 1796047] Re: update-ieee-data throws error because of wrong url

2019-10-22 Thread Oliver Grawert via ubuntu-bugs
** Also affects: ieee-data (Ubuntu Xenial) Importance: Undecided Status: New -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1796047 Title: update-ieee-data throws error because of wrong

[Bug 1844498] Re: 18.10+ cloud images have the LXD group as gid 1000

2019-10-10 Thread Oliver Grawert via ubuntu-bugs
note that the build scripts for core do have an md5sum check to detect changes to /etc/passwd|group|shadow because this file is orignally readonly. dirs created with specific UIDs/GIDs by package postinist scripts that get copied into the writable area of the rootfs during first boot would fail

[Bug 1866002] [NEW] ubuntu-image should allow to disable console-conf at build time

2020-03-04 Thread Oliver Grawert via ubuntu-bugs
Public bug reported: currently various odd hacks exist to make sure that console-conf does not start up on appliance or kiosk systems that are supposed to be user- less and get managed only via the snapd REST API. to disable console-conf starting up the file /var/lib/console- conf/complete must

[Bug 1866058] [NEW] gadget hooks should not write to /root/snap by default

2020-03-04 Thread Oliver Grawert via ubuntu-bugs
Public bug reported: using a qemu image I recently noticed that there is a /root/snap/pc dir on the system ... the pc gadget ships an empty no-op configure hook that apparently gets run during seeding. since snapd knows what a gadget is, it would be nice if gadget hooks could by default made to

[Bug 1866193] Re: error: too early for operation, device not yet seeded or device model not acknowledged

2020-03-06 Thread Oliver Grawert via ubuntu-bugs
*** This bug is a duplicate of bug 1864252 *** https://bugs.launchpad.net/bugs/1864252 ** This bug has been marked a duplicate of bug 1864252 preseeded snap installs fail in images -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to

[Bug 1864162] Re: Don't install snap programs.

2020-02-22 Thread Oliver Grawert via ubuntu-bugs
*** This bug is a duplicate of bug 1864252 *** https://bugs.launchpad.net/bugs/1864252 ** This bug has been marked a duplicate of bug 1864252 preseeded snap installs fail in images -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to

[Bug 1863214] Re: Add storage kernel modules for arm64/armhf build

2020-02-14 Thread Oliver Grawert via ubuntu-bugs
The pc-kernel snap is essentially just "linux-generic", sadly it was decided back then that the "linux-generic" name was not appropriate and that we would only build the kernel snap for amd64 in core16 In core 18 it was finally accepted to also build this kernel for both arm arches so that

[Bug 1873797] Re: add_mountroot_fail_hook is either buggy or not used properly, causes boot problems

2020-04-20 Thread Oliver Grawert via ubuntu-bugs
also ... i highly doubt that support for lvm2 is planned in UbuntuCore 20, this bug might rather be for the initramfs-tools package itself ... -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1873797

[Bug 1872065] Re: 'writable' should not be introduced as a new label for persistence partition in Ubuntu 20.04

2020-04-20 Thread Oliver Grawert via ubuntu-bugs
side-note: /writable was introduced in 2014 FWIW -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1872065 Title: 'writable' should not be introduced as a new label for persistence partition in

[Bug 1872736] Re: install from USB to USB device wipes grub setup on NVME disk

2020-04-26 Thread Oliver Grawert via ubuntu-bugs
Sorry for the late reply, here is my /var/log/installer/syslog ** Attachment added: "/var/log/installer/syslog" https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/1872736/+attachment/5360826/+files/syslog -- You received this bug notification because you are a member of Ubuntu Bugs,

[Bug 1575053] Re: Please move the "$HOME/snap" directory to a less obtrusive location

2020-04-30 Thread Oliver Grawert via ubuntu-bugs
To extend what zyga said above ... A major amount of snaps are used on appliance, embedded, IoT and server devices (in fact snaps were used in these environments for two years already before they came to the desktop at all). On many of these systemd XDG doesnt even remotely play a role so it was

[Bug 1872736] [NEW] install from USB to USB device wipes grub setup on NVME disk

2020-04-14 Thread Oliver Grawert via ubuntu-bugs
Public bug reported: For a system test I prepared a USB stick with the current focal iso, booted from it and attempted an install to another (identical) USB disk. During partitioning /dev/sdb (the USB disk) was picked as target for the bootloader installation, a 63GB ext4 partition for / and a

[Bug 1872736] Re: install from USB to USB device wipes grub setup on NVME disk

2020-04-14 Thread Oliver Grawert via ubuntu-bugs
Note that the install to USB seems to have done what it should just fine ... ...but that it wiped existing the bootloader on the internal disk that was explicitly de-selected everywhere is not okay. -- You received this bug notification because you are a member of Ubuntu Bugs, which is

[Bug 1878449] Re: 20.04's desktop installer always fsck's when starting

2020-05-13 Thread Oliver Grawert via ubuntu-bugs
this is intentional and you can skip it with ctrl-c https://discourse.ubuntu.com/t/defaulting-to-verify-the-image-integrity- before-installing-on-desktop/13472 -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu.

[Bug 1877852] [NEW] Evolution not showing unread count in dock, missing build-dep

2020-05-10 Thread Oliver Grawert via ubuntu-bugs
Public bug reported: Evolution has the ability to show the unread messages count in the Dock app icon, this works for the Ubuntu Dock GNOME extension as well as for Dash to Dock and indeed the old unity7 dock ... simply re-compiling the current focal evolution package with an added libunity-dev

[Bug 1877852] Re: Evolution not showing unread count in dock, missing build-dep

2020-05-10 Thread Oliver Grawert via ubuntu-bugs
** Attachment added: "evo-counter.png" https://bugs.launchpad.net/ubuntu/+source/evolution/+bug/1877852/+attachment/5369430/+files/evo-counter.png -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu.

[Bug 1575053] Re: Please move the "$HOME/snap" directory to a less obtrusive location

2020-05-08 Thread Oliver Grawert via ubuntu-bugs
to move the content of the dir you might be able to: stop all snap apps ... mv ~/snap /whereever/your/bigger/disk/is/mounted mkdir ~/snap sudo mount --bind /whereever/your/bigger/disk/is/mounted/snap ~/snap you might need to adjust permissions on the target disk, the snap dir must be fully

[Bug 1877852] Re: Evolution not showing unread count in dock, missing build-dep

2020-05-11 Thread Oliver Grawert via ubuntu-bugs
seems installing evolution-indicator solves it without re-compiling ... closing the bug as invalid ... ** Changed in: evolution (Ubuntu) Status: New => Invalid -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu.

[Bug 1875493] Re: [core] log rotation doesn't properly restart rsyslogd

2020-05-14 Thread Oliver Grawert via ubuntu-bugs
it is a hard requirement of the SRU process that fixes land in the devel release and then propagate backwards through all other releases ... -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1875493

[Bug 1873797] Re: add_mountroot_fail_hook is either buggy or not used properly, causes boot problems

2020-05-13 Thread Oliver Grawert via ubuntu-bugs
pleas use 1878378 for further discussion of that issue, there is no relation at all to initramfs-tools-ubuntu-core ** Changed in: initramfs-tools-ubuntu-core (Ubuntu) Status: Incomplete => Invalid -- You received this bug notification because you are a member of Ubuntu Bugs, which is

[Bug 1876862] Re: Missing v3d driver disables 3D support on RPi4

2020-05-13 Thread Oliver Grawert via ubuntu-bugs
adding bionic since we're supposed to fully support the pi4 there too ** Also affects: linux-raspi (Ubuntu Bionic) Importance: Undecided Status: New -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu.

[Bug 1875493] Re: [core] log rotation doesn't properly restart rsyslogd

2020-05-08 Thread Oliver Grawert via ubuntu-bugs
well, rsyslog is completely gone from core after UC16 ... how would you test that in UC18 or UC20 to verify SRUs ? -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1875493 Title: [core] log rotation