missing a && between icicle and visionfive, led to /boot/efi still being
in place, and grub-install running instead of exiting the func.
fixes LP:2015750
Cloud-init cannot write directly to
/etc/NetworkManager/system-connections because subiquity may
need to emit config to /etc/netplan/00-installer.yaml and call
netplan apply for autoinstall.network use-cases.
When cloud-init's config is written directly to
/etc/NetworkManager, neither netplan nor subiquity has knowledge of
this config and this results in namespace collisions in NetworkManager
due to `netplan-` named connections and `cloud-init` connection ids
fighting over which config own a given interface name.
Deleting this config overlay allows subiquity to manage all network
setup when it needs to with netplan directly.
Subiquity already has logic to rename any unwanted netplan
configuration when it intends to write cfg and run netplan apply[1].
This should allow subiquity full control of network config when needed.
[1] https://github.com/canonical/subiquity/blob/
92ac6544cdfedfd332d8cd94dbcfad0aab994575/subiquitycore/
controllers/network.py#L267
LP: #2015605
SUBARCH=visionfive2 is used to build images for the StarFive VisionFive 2
boards. For the device-tree we assume board revision 1.3B.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Autoinstall directives can be provided on the grub cmdline to
cloud-init via kernel parameters like the following:
autoinstall 'ds=nocloud-net;s=http://somedomain/'
In order to support DNS resolution for NoCloud datasource at
datasource discovery time, cloud-init.service needs to be
orderered after NetworkManager.service and
NetworkManager-wait-online.service
which will have brought up applicable NICs.
Since NetworkManager is After=dbus.service, the cloud-init.service
avoids systemd ordering cycles by also dropping
Before=sysinit.target when it adds, After=NetworkManager.service and
After=NetworkManager-wait-online.service
Add this file overlay for /lib/systemd/system/cloud-init.service
because systemd drop-in files can only add constraints and not
drop prexisting service constraints.
Also add an AUTOMATION_HEADER comment to any generated files to
add discoverability in the event of future bugs/concerns.
LP: #2008952
ipc was dropped as an apparmor feature. checked by grabbing the latest
lunar VM, installing the latest kernel, doing a reboot, and comparing
directories and files. compared all files and the only diff is the ipc
posix_mqueue
When building locally using the auto/build script unmounting fails.
Avoid mounting via bind. Mount mountpoint/dev as devtmpfs file system and
mountpoint/dev/pts as devpts file system.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Evolve the seed to only ship the specific part useful to WSL users. This
allows to trim down the image size.
Co-authored-by: Jean-Baptiste Lallement <jean-baptiste@ubuntu.com>
For the SiFive HiFive Unmatched board we create a pre-installed image using
u-boot-menu. Increase the watchdog threshold in this case too.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
With Radeon GPUs and kernel 5.19 a soft lockup was observed.
Increase the watchdog threshold.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Using numbered configuration fragments makes the order of application
easier to track
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
This allows to consolidate linux-kvm and linux-generic kernel
flavours. This brings the perfomance benefit of linux-kvm flavour to
all cloud and pre-installed images. It does trade data-safety.
LP: #2006511
Signed-off-by: Dimitri John Ledkov <dimitri.ledkov@canonical.com>
According to the EBBR specification the GPT partitions for firmware should
have attribute bit 0 (Required Partition) set.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Since version 2022.10 U-Boot SPL and U-Boot are installed onto the same partition.
Package nezha-boot0 is not needed anymore.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
RISC-V boards tend to boot slowly.
We should provide progress information when booting.
Use 'efi=debug earlycon' on the Linux command line via new file
/etc/default/grub.d/cmdline.cfg.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
We add a ubuntu user inside the image because we
want to have a operational nonroot user and also
be aligned with the other Ubuntu images.
Signed-off-by: Samir Akarioh <samir.akarioh@canonical.com>
The Nezha and the LicheeRV boards do not have enough memory for an initrd
with most modules. Therefore the number of included modules has to be
reduced.
Create file /etc/initramfs-tools/conf.d/modules_list.conf
to set MODULES=list.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Remove redirections of type
command &1>2
Executing the command in the background and creating and empty file '2'
was never intended.
As the messages are information only redirecting to stderr would not make
sense either.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
So that people without network access can download the package and
install it using a usb drive for example.
Signed-off-by: Alexandre Ghiti <alexandre.ghiti@canonical.com>
This reverts commit d5a7d6655f.
The Wifi driver package is in universe and can't be promoted in time for
the release, so revert this.
Signed-off-by: Alexandre Ghiti <alexandre.ghiti@canonical.com>
The LicheeRV Dock board comes with only 512MB of DRAM so the only difference
with a Nezha image is the fact that we have to remove
cryptsetup-initramfs package which makes the initrd too big for the
board to boot.
Signed-off-by: Alexandre Ghiti <alexandre.ghiti@canonical.com>
Current Kinetic GCE image builds are failing with the following error:
update-initramfs: Generating /boot/initrd.img-5.19.0-1004-gcp
zstd: error 25 : Write error : No space left on device (cannot write compressed block)
E: mkinitramfs failure zstd -q -1 -T0 25
Seems like after `linux-gcp` update from 5.15 to 5.19 `linux-modules` package
has gotten ~40MB larger and with that GCE image builds are over the edge wrt
available disk space in chroot.
Bumped up disk image size for amd64 to 3.5GB to match the sizes used by armhf
and generic images.
The cloud-init bug (see LP:1968873) got fixed now so using a sshd
config snippet should work now.
This partly reverts commit aa1be5eaaa
but uses now 60-cloudimg-settings.conf instead of
10-cloudimg-settings.conf .
For now, all RISC-V hardware is SBC-like board which embed a Wifi
chipset so install wpasupplicant by default. We'll certainly split the
seeds between server and embedded hardware later.
Signed-off-by: Alexandre Ghiti <alexandre.ghiti@canonical.com>
* Fix some issues with the netboot tarballs:
- Include the signed shim (oops).
- Make the kernel path on disk and in the bootloader config match (more
oops).
- Make paths more architecture dependent as the code in grubnetXXX.efi to
probe a platform dependent path first doesn't work.
While merging the VisionFive support, we removed the installation of
u-boot-menu for the Unmatched by mistake: fix this by reinstating it.
Fixes: ce9f5cacca ("riscv: Add support for StarFive VisionFive")
Signed-off-by: Alexandre Ghiti <alexandre.ghiti@canonical.com>
This change triggered a bug in cloud-init (see LP:1968873). cloud-init
does not recongnize sshd options set in /etc/ssh/sshd_config.d/ and
cloud-init modifies directly /etc/ssh/sshd_config which gets then
overwritten by settings from /etc/ssh/sshd_config.d/ .
This reverts commit b54d24ff33.
3.5G is not enough for riscv64 preinstalled as the creation of the initrd fails
with the following error:
Creating config file /etc/default/grub with new version
Processing triggers for initramfs-tools (0.140ubuntu13) ...
update-initramfs: Generating /boot/initrd.img-5.15.0-1011-generic
zstd: error 25 : Write error : No space left on device (cannot write compressed block)
E: mkinitramfs failure zstd -q -1 -T0 25
update-initramfs: failed for /boot/initrd.img-5.15.0-1011-generic with 1.
dpkg: error processing package initramfs-tools (--configure):
installed initramfs-tools package post-installation script subprocess returned error exit status 1
Signed-off-by: Alexandre Ghiti <alexandre.ghiti@canonical.com>
The image created uses a UEFI bootflow, so we install grub for this board
only. We also need flash-kernel to install the dtb where grub can find
it.
This image is specifically architectured so that it can be installed on
a "factory" board, meaning using the u-boot firmware which was
originally implemented for Fedora, so we need the p3 partition that
embeds a uEnv.txt file to tell u-boot what/where to load next stage.
Signed-off-by: Alexandre Ghiti <alexandre.ghiti@canonical.com>
Define the image layout for the Nezha board.
The U-Boot SPL based boot0 may be installed starting in sector 16 or 256.
As sector 16 is incompatible with GPT partitioning use sector 256.
The primary U-Boot image is expected to start at sector 32800 and its
backup in sector 24576.
Cf. https://linux-sunxi.org/index.php?title=Allwinner_Nezha&oldid=24469
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Modifying directly /etc/ssh/sshd_config creates "problems" when
upgrading eg. from Focal to Jammy because the upgrade will ask the
user what to do with the modified config. To avoid that, put the
custom configuration into /etc/ssh/sshd_config.d/ so the upgrade of
openssh-server can just replace /etc/ssh/sshd_config without asking
the user.
This reverts part of a change causing regression with vmware import due to the
cdrom getting moved to SCSI while shifting controller IDs. (LP: #1970795)
Germinate doesn't take very long at all to run but downloading the
indices it operates on can take a while and nothing else in auto/config
does so not doing it every time you run "lb config" can be a real time
saver.
The code that invokes germinate already checked if the output was
already there but it was unconditionally deleted by the time control got
to that point.
LP: 1969664 tracks an issue related to the deprecation of rsa+ssh on
Jammy+ openssh server, coupled with upstream vagrant bugs, that cause
Jammy vagrant images fail to bootstrap due to ssh negotiation issues.
Moving to a different key algo from the upstream insecure key matches
Jammy's expectations, and works with older vagrant versions.
vagrant >= 2.2.16 hosts are unaffected by the issue, as an upstream
change was made. This change keep compatibility with newer vagrant
versions as well.
Readding this file per reviewer's request until CPC splits the
pipelines. Removing this file would make CPC image builds fail.
Co-authored-by: Didier Roche <didrocks@ubuntu.com>
Commit 245f7772bd added code to abort the build if a snap wants to
install "core" (the 16.04 runtime). That's great but there are still
some CPC maintained image builds that use snaps based on "core". So
make it possible to continue the build if the "ALLOW_CORE_SNAP" env
variable is set.
Due to how `disk-image` file is structured, it builds BIOS and UEFI
images at the same time. However, certain images (e.g., GCE images)
require only UEFI image to be built, BIOS image is being simply
discarded. This results in longer build times.
Splitting out `disk-image-uefi` would allow images to use it instead of
`disk-image` and thus avoid building unused BIOS images.
`disk-image` now depends on `disk-image-uefi` for backward
compatibility.
The UNCONFIGURED FSTAB warning was being left in the result, the discard
option wasn't included, and the fsck flag was 0 (all in marked contrast
to the preinstalled server images).
Changes in either livecd-rootfs or ubuntu-image seem to periodically
break the transfer of the pre-allocated swapfile (copying it in such a
fashion that it winds up "with holes" and thus unable to be used as a
swapfile). Rather than fight this, just use a simple systemd service to
generate the swapfile if it doesn't exist (using fallocate to keep
things snappy).
This fixes GCE shielded VM instances integrity monitoring failures on
focal and later. Our images are built with an empty /boot/grub/grubenv
file, however after the first boot `initrdless_boot_fallback_triggered`
is set to 0. This change in `grubenv` results in integrity monitoring
`lateBootReportEvent` error.
It seems that the only thing that's checking for this `grubenv` variable
is `grub-common.service`, and it is looking specifically for a `1`
value:
if grub-editenv /boot/grub/grubenv list | grep -q
initrdless_boot_fallback_triggered=1; then echo "grub:
GRUB_FORCE_PARTUUID set, initrdless boot paniced, fallback triggered.";
fi
Unsetting this variable instead of setting it to 0 would prevent issues
with integrity monitoring.
LP: 1960537 illustrates an issue where the calls to e2fsck in the
umount_partition call are failing due to an open file handle. At this
time, we are unable to find a root cause, and it's causing many builds
to fail for CPC. Adding a sleep 30 as a workaround as the file handle
releases within that timeframe. This does not address root cause.
Currently the RISC-V preinstalled server images come with partitions that
are only 1 KiB aligned. Ext4 may use 4 KiB block size. The existing
misalignment leads to decreased performance.
Decrease the size of the loader2 partition by 34 512-byte blocks. This
results in 1 MiB alignment of the EFI and root partitions.
The remaining loader2 partition size of close to 4 MiB is still large
enough for U-Boot or a future EDK II.
Fixes: a808b28d47 ("riscv64: build preinstalled riscv64 image with uboot SPL and CIDATA.")
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
livecd-rootfs creates non-private mounts. When building locally using
the auto/build script unmounting fails.
To unmount dev/pts it is insufficient to make the mount private. Its
parents must be private too. Change teardown_mountpoint() accordingly.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Current jammy builds fail with:
dpkg: error processing archive /var/cache/\
apt/archives/grub-common_2.04-1ubuntu48_armhf.deb (--unpack):
cannot copy extracted data for './usr/share/grub/unicode.pf2' \
to '/usr/share/grub/unicode.pf2.dpkg-new': \
failed to write (No space left on device)
It hangs during booting when upgrading hardware
version ESXi after deploying image in groovy.
(Current default version is 10)
It could be resolved by adding serial port in VM
when vm version is larger than 10.
Seriaol port1 has been configured as default so
we need to change setting serial0 as false.
As wsl is an image target of ubuntu-cpc, the base seed is hardcoded to
ubuntu-server instead of wsl one. For now, add it, as for the other
cpc images, in hooks.
LP: 1944004 described an issue where a libc transition caused snapd
seccomp profiles to reference a path that no longer existed, leading to
permission denied errors. The committed fix for snapd then raised an
issue where running `snapd debug seeding` would present a
preseed-system-key and seed-restart-system-key due to a mismatch
between the running kernel capabilities and the profiles being loaded by
snapd. By mounting a cgroup2 type to /sys/fs/cgroup, the capabilities
match for snapd as mounted in the chroot. This is done similarly to
live-build/functions:138-140 where apparmour and seccomp actions are
mounted after updating the buildd.
Debian changelog.Debian.* files are already keept for minimized
builds. But those changelogs are from non-native .deb packages (see
man dh_installchangelogs). Native .deb packages name their changelog
just changelog.* . So keep them in a minimized build, too.
LP: #1943114
otherwise each and every layer above a layer with a kernel gets its own
initramfs, which is silly.
Copy/paste the cruft cleaning bit of lb_chroot_hacks to be run on
non-live layers.
for the live server build, i want to make a layer to install the kernel
into but do not want the layer itself to be published.
the implementation is a bit clunky but it works.
At this point all of the custom final_message is now obsolete.
Remove it, letting us instead use the default final_message.
Leave a note about the above.
groovy hangs during boot on ESXi when the version is greater than
10. Adding a serial port by default fixes this specific bug - increasing
the HW version will be for another branch.
This is because more investigation is needed into whether it is possible to
increment ddb.virtualHWVersion without disrupting Oracle VirtualBox images.
Some packages are in universe at release time then promoted to
the main pocket in -updates during the release lifecycle.
These packages should be considered by germinate when the root fs is
built (LP: #1921862)
Co-authored-by: Didier Roche <didrocks@ubuntu.com>
Initialize passwords from sources.list.
Use urllib everywhere.
This way authentication is added to all the required requests.
And incoming headers, are passed to the outgoing requests.
And all the response headers, are passed to the original client.
And all the TCP & HTTP errors are passed back to the client.
Thus should avoiding hanging requests upon failure.
Also rewrite the URI when requesting things.
This allows to use private-ppa.buildd outside of launchpad.
Signed-off-by: Dimitri John Ledkov <xnox@ubuntu.com>
armhf & arm64 images use grub. And despite disk-image &
disk-image-uefi installing all the grubs, some of the configuration is
done in the 999-cpc-fixes. Specifically removal of "quiet splash" is
done there, but not active on armhf & arm64. This results in arm
images to boot with "quiet splash".
Enable running the later portions of 999-cpc-fixes on armhf & arm64.
Drop duplicate call to update-grub, as update-grub2 is simply a
symlink to update-grub.
Add a guard around the call to reconfigure grub-pc, to only do that
when it is installed.
This makes armhf & arm64 uefi images consistent with amd64 uefi
images.
LP: #1925780
With that, the Dockerfile modifications[0] currently done externally
are done now here. That means that the created rootfs tarball can be
directly used within a Dockerfile to create a container from scratch:
FROM scratch
ADD livecd.ubuntu-oci.rootfs.tar.gz /
CMD ["/bin/bash"]
[0]
https://github.com/tianon/docker-brew-ubuntu-core/blob/master/update.sh
This is a copy of the ubuntu-base project.
Currently ubuntu-base is used as a base for the docker/OCI container
images. The rootfs tarball that is created with ubuntu-base is
published under [0]. That tarball is used in the FROM statement of the
Dockerfile as base and then a couple of modifications are done inside
of the Dockerfile[1].
The ubuntu-oci project will include the changes that are currently
done in the Dockerfile. With that:
1) a Dockerfile using that tarball will be just a 2 line thing:
FROM scratch
ADD ubuntu-hirsute-core-cloudimg-amd64-root.tar.gz /
CMD ["/bin/bash"]
2) Ubuntu has the full control about the build process of the
docker/OCI container. No external sources (like [1]) need to be
modified anymore.
3) Ubuntu can publish containers without depending on the official
dockerhub containers[2]. Currently the containers for the AWS ECR
registry[3] use as a base[4] the official dockerhub containers. That's
no longer needed because a container just needs a Dockerfile described
in 1)
When the ubuntu-oci project has the modifications from [1] included,
we'll also update [1] to use the ubuntu-oci rootfs tarball as a base
and drop the modifications done at [1].
Note: Creating a new ubuntu-oci project instead of using ubuntu-base
will make sure that we don't break users who are currently using
ubuntu-base rootfs tarballs for doing their own thing.
[0] https://partner-images.canonical.com/core/
[1]
https://github.com/tianon/docker-brew-ubuntu-core/blob/master/update.sh
[2] https://hub.docker.com/_/ubuntu
[3] https://gallery.ecr.aws/ubuntu/ubuntu
[4]
https://launchpad.net/~ubuntu-docker-images/ubuntu-docker-images/+oci/ubuntu/+recipe/ubuntu-20.04
One can call divert_grub; replace_kernel; undivert_grub. And
replace_kernel will call into force_boot_without_initramfs, which
under certain conditions can call divert_grub &
undivert_grub. Resulting in undivert_grub called twice in a row.
When undivert_grub is called twice in a row it wipes
systemd-detect-virt binary from disk, as the rm call is unguarded to
check that there is something to divert if systemd package is
installed. And if the systemd package is not installed, it does not
check that systemd-detect-virt file is in-fact what divert_grub has
created.
Add a guard to check that systemd-detect-virt is the placeholder one,
before removing it.
LP: #1902260
The case is for arch:subarch combo, not just arch alone even if
subarch is empty. Thus currently on adm64/arm64/armhf ubuntu-cpc
builds mbr image is created and then ignored, as the convert to qcow2
hook prefers the uefi image whenever available.
Skipping building these correctly, should speed up the build a little
bit and use slightly less disk space.
shim-signed depends on grub-efi-amd64-signed, which in turn has
alternative depends on either `grub-efi-amd64 | grub-pc`. However to
support booting with either via shim&signed-grub and BIOS, the choice
must be made to install grub-pc, not grub-efi-amd64.
This makes images consistent with Ubuntu Deskop, Live Server, buildd
bootable images; all of which already do install grub-pc and
shim-signed.
LP: #1901906
vmtools version in vmdk header (LP: #1893898)
LP: #1893898 describes missing vmtools version from the vmdk headers.
The version should be added as ddb.toolsVersion = "2147483647" however
the sed was no longer replacing a ddb.comment field with the tools
version. Rather than subbing ddb.comment with toolsVersion, this commit
deletes ddb.comment (which the comment mentions could cause errors),
and adds the correct value. There was no visibility into the descriptor
during hook creation, so debug statements were added. This allows us to
quickly verify in the logs that bad statements are removed (the possibly
offending commetns), as well as ensuring that the toolsVersion is added
MP: https://code.launchpad.net/~jchittum/livecd-rootfs/+git/livecd-rootfs/+merge/392401
When desktop-preinstalled image options were added in
38157b3748, for the raspi subarch, the
options listed there were not scoped for raspi subarch. This results
in those options getting also applied for the HYPERV
ubuntu:desktop-preinstalled image.
Thus scope the newly added options under raspi subarch case only.
Regression introduced in 38157b3748 when
desktop-preinstalled code branch was added, it dropped addint
ubuntu-desktop task. Instead it added ubuntu-desktop-raspi task, only
for the raspi subarch, which depends on ubuntu-desktop. But the hyperv
case, now ended up without ubuntu-desktop task.
It looks like introduction of "desktop-preinstalled" assumed, that it
is for raspi only, when in fact that code path now started to be used
for hyperv gallery image too.
The CPC build hooks for amd64 incorrectly attempt to install shim-signed
in addition to grub-efi-amd64 and grub-pc. These latter two packages
conflict with each other. Instead shim-signed should install whatever
packages are required.
Additionally, this will ensure that autoremove is run after installing
anything in the CPC build hooks. This is done to avoid shipping images
that include packages that are autoremovable. This will clean-up as
packages are installed and detect any breakage at build time.
Multipass on Mac OS X requires standalone kernel and initrd artifacts
to boot.
Also call update-initramfs on all installed kernels. We only have one
kernel installed, so we don't need to specify an explicit version.
There was a question on if the comment removals in the `sed` were
required. The comments (`#`) are created by vmdk-stream-converter and
seem to cause no issues. `ddb.comment` is no longer being written by the
tool anymore. Moved the check earlier to ensure the new header isn't too
large before running truncate (otherwise it may be too long, and we
remove bits we want)
LP: #1893898 describes missing vmtools version from the vmdk headers.
The version should be added as ddb.toolsVersion = "2147483647" however
the sed was no longer replacing a ddb.comment field with the tools
version. Rather than subbing ddb.comment with toolsVersion, this commit
deletes ddb.comment (which the comment mentions could cause errors),
and adds the correct value. There was no visibility into the descriptor
during hook creation, so debug statements were added. This allows us to
quickly verify in the logs that bad statements are removed (the possibly
offending commetns), as well as ensuring that the toolsVersion is added
virtualbox-guest-utils kernel modules is included in linux-modules
starting in kernel 5.4.0-33 in focal-updates. The vagrant hook also
explicit installed virtualbox-guest-utils. An error occurred with the
version installed from the archives, however, with the inclusion in
linux-modules, there's no need to explicitly install
virtualbox-guest-utils. Removes the code for the explicit install.
MOUNTPOINT_BACKUP_SOURCE_LIST is exposed when you call
setup_mountpoint. Consumers can use this variable if they need to
explicitly change something in sources.list wihout relying on the name
livecd-rootfs chooses.
Original fix proposed by Stanislav German-Evtushenko (giner)
CPC Ubuntu cloud images default to enabling a serial console connection
via the kernel commandline option `console=ttyS0`. Many clouds support
the serial connection, and utilize it for debugging purposes. Virtualbox
supports the serial connection as well. In Bionic and earlier images,
Vagrant boxes created a serial log file in the directory of the
Vagrantfile by default. However this is not standard behaviour for
Vagrant images, and so it was removed in Eoan onwards.
Starting in Eoan, there were reports of image booting slowdown (1874453
is a single example). After testing, it was determined that the serial
connection starting, without a device attached, was the cause of the
slow down. However, we did not want to revert to the old functionality
of creating a file. Much thanks to <giner> for providing the Ruby syntax
for sending to File::NULL.
This option will not create a local file, however, the default
Vagrantfile configuration is overwritable via a users Vagrantfile. The
original syntax for creating a file local to the users Vagrantfile has
been included as an example.
These introduced a regression for ppc64el and needs more time to bake.
This reverts commits 1deb0c68e8 &
6dbb30f53b.
* "ubuntu-cpc: Fix ppc64el grub console update"
* "ubuntu-cpc: Disable boot splash in all cloud images (LP: #1725358)"
The commit 6dbb30f5 (2.682) which disabled boot splash for all cloud
images introduced an error in the ppc64el hook. This patch corrects the
name of the variable that contains grub console overrides. The error
seen during testing was
'disk-image-ppc64el.binary: line 44: CONSOLES: unbound variable'
and this was due to a typo.
When trying to debug an issue on ARM64 it was reported that it was
quite difficult to debug because of control codes on the console from
the splash.
For cloud image there is a chroot customization the drops 'quiet splash'
but this is only applied to amd64. It hasn't made it into other
architectures because they don't have grub by default in the chroot.
However, when we get into binary hook for the uefi disk image and it's
derivatives grub is installed and this includes architectures that were
skipped in the chroot hook.
This patch changes the cpc-fixes chroot hook to add a cloud-images
grub config with basic overrides, including dropping the boot splash,
for all architectures. For images that never get grub installed this
addition is harmless and small while ensuring that the grub experience
is consistent for images that have grub. The configuration of console
devices as hard-coded remains arch specific.
The seed now specifies the lxd snap in focal as
'lxd=4.0/stable/ubuntu-20.04' which doesn't match the expectations of
the code with looks for lxd as the only snap in the seed for minimized
images. This patch updates the pattern to accept 'lxd' or 'lxd=*'.
In v2.672 the default boot behavior of cloud images changed:
- Prior to v2.672, cloud images with the linux-generic kernel attempt
to boot without an initramfs, would fail, and then retry with an
initramfs.
- After v2.672, cloud images with the linux-generic kernel boot with
an initramfs on the first try.
While the behavior is different between the two, they both result in
an instance that has booted with an initramfs. To ensure the changes
in v2.672 do not regress, we need an automated way to check if we are
attempting to boot without an initramfs and failing.
With this change, when we attempt to boot with an initramfs and fail,
initrdless_boot_fallback_triggered is set to non-zero in the grubenv.
This value can be checked after boot by looking in /boot/grub/grubenv
or by using the grub-editenv list command.
Builds in LP with the Xenial kernel were happy with the recursive mount of
/sys inside the chroot while performing snap-preseeding but autopkgtests
with the groovy kernel failed. With the groovy kernel the build was
unable to unmount sys/kernel/slab/*/cgroup/* (Operation not permitted).
This patch mounts /sys and /sys/kernel/security in the chroot in the
same way we've added for binary hooks. This provides the paths under
/sys needed for snap-preseed while avoiding issues unmounting other
paths.
The snap-preseed command can do a number of things during the build
that are currently performed at first boot (apparmor profiles, systemd
unit generation, etc). This patch adds a call to reset the seeding and
apply these optimizations when adding a seeded snap. As a prerequisite
to calling snap-preseed we need to make /dev/mem available as well as
mounts from the host to perform this work, so those are also added here.
I recently pulled initramfs logic out of the base build hook, and
dropped that into the `replace_kernel` function. Any cloud image that
does not leverage the generic virtual kernel was expected to call
`replace_kernel` to pull in a custom kernel. That function will
disable initramfs boot for images that use a custom kernel.
Minimal cloud images on amd64 use the linux-kvm kernel, but the build
hook does not utilize the `replace_kernel` function. Instead, the
kernel flavor is set in `auto/config`. I pulled that logic out of
`auto/config` and am now calling `replace_kernel` in the build hook.
I also moved a call to generate the package list so that it will pick
up the change to the linux-kvm kernel.
snap_name[/classic]=track/risk/branch is now the supported snap name
specification, which allows to specify the full default track and
optional classic confinemnt.
Supporting such specification in the seedtext allows one to specify a
better default channel. For example, this will allow lxd to switch
from latest/stable/ubuntu-20.04 to 4.0/stable/ubuntu-20.04 as 4.0 is
the LTS track matching 20.04 support timeframe.
LP: #1882374
Initramfs-less boot, which is a boot optimization, should only be
applied where we know it could work for users and provide an improved
boot boot experience; images with custom kernels are candidates for
that.
Generic cloud images with the linux-generic kernel are not able to
boot without an initramfs. Previously, these images attempted to boot
without an initramfs, would fail, and then retry with an initramfs.
This slows the boot and is confusing behavior.
It was reported and confirmed in LP bug #1875400
(https://bugs.launchpad.net/cloud-images/+bug/1875400) that on the public
KVM cloud image there exists a large list of packages marked for auto-removal.
This should never be the case on a released cloud image.
These packages are marked for auto-removal because in the KVM image binary hook
we removed both initramfs-tools and busybox-initramfs packages. Due to package
dependencies this also removed:
busybox-initramfs* cloud-initramfs-copymods* cloud-initramfs-dyn-netconf*
cryptsetup-initramfs* initramfs-tools* initramfs-tools-core* multipath-tools*
overlayroot* sg3-utils-udev* ubuntu-server*
But it did not remove all the packages that the above list depended on.
This resulted in all those packages being marked for auto-removal because they
were not manually installed nor did they have any manually installed packages
that depended on them.
The removal of initramfs-tools and busybox-initramfs was to avoid the
generation of initramfs in images that should boot initramfsless.
This requirement is obsolete now because the initramfsless boot handling
is now handled via setting GRUB_FORCE_PARTUUID in /etc/default/grub.d/40-force-partuuid.cfg.
In test images I have verified that GRUB_FORCE_PARTUUID is set and that
boot speeds have not regressed.
LP: #1875400
Vagrant images were previously put at 10G, but this was a regression
from Trusty, in which they were 40G. This made it a tough sell for
users to upgrade if they were using a Ubuntu desktop experience.
This change does not impact disk usage as Vagrant with the virtualbox
provider dynamically allocates space with the VMDK. On a test system,
the VMDK took up 1.1G of disk space according to df, and after
creating a 2G file in Vagrant, the VMDK grew to 3.1G.
Therefore, users who are running on a system with little free space will
not see adverse effects if they upgrade to a new vagrant image
Seeing any snap via snap_preseed will evaluate the base for each snap
and seed the appropriate base. There should be no reason to explicitly
seed the 'core' snap and with snaps moving to 'core18' this will add
'core' without need.
The _snap_post_process function is meant to install snapd if core18 is the
only core snap installed or removed snapd if core is installed and snapd
was not explicitly installed. But the current logic in _snap_preseed
will never call _snap_post_process. $core_name will never be empty
with the existing logic, but even if it were that would only be for the
'core' snap and we'd miss using the 'core18' logic that pulls in snapd.
Given the case statement in _snap_post_process can handle doing the
right thing given any snap we can just call it unconditionally.
In the buildd image chroot, /etc/resolv.conf is a symbolic link to
a configuration file in the /run directory. A call to truncate will
modify that file, which we should not do. Instead, we want to remove
the symbolic link and replace it with an empty file.
The base image is built with packages from the release pocket; however,
we want the latest from updates and security. Those pockets are already
enabled, we just need to perform an upgrade to pull in the latest
packages.
Back in 2017 some code was added to ignore failures tearing down loop
devices. But debugging that growpart race on cloud images made me (very)
aware of a potential cause of the race: doing something like zerofree on
a device will cause udev scripts to run, and if they are still running
by the time kpartx is called, you would expect the kpartx -d to fail. So
lets see if a udevadm settle helps, and get rid of one of the "sometimes
this fails but we don't know why" comments...
Łukasz 'sil2100' Zemczak
Steve Langasek
John Chittum
Chad Smith
Heinrich Schuchardt
William Wilson
Dave Jones
Dan Bungert
Dimitri John Ledkov
Sean Davis
Michael Hudson-Doyle
Didier Roche
Erich Eickmeyer
Sebastien Bacher
Utkarsh Gupta
Samir Akarioh
Brian Murray
Simon Poirier
Alexandre Ghiti
Ivan Kapelyukhin
Rudra Bali
Thomas Bechtold
Michał Sawicz
Jean-Baptiste Lallement
Jess Jang
Didier Roche
Jason C. McDonald
Ethan Hsieh
Chloé Smith
Dimitri John Ledkov
Gauthier Jolly
Cody Shepherd
Iain Lane
Jinming Wu, Patrick
Robert C Jennings
Joshua Powers
David Krauser
Pat Viafore
Stanislav German-Evtushenko
Balint Reczey
Philip Roche