Commit [1] introduced an Armv7 test case for Xen named TestXenArmv7,
next to the original Aarch64 test case which was simply named TestXen.
The test list shows, for example with the command
"support/testing/run-tests -l":
tests.package.test_xen.TestXen
tests.package.test_xen.TestXenArmv7
In order to make this test list a bit more explicit, this commit
renames the TestXen to TestXenAarch64. With that change, the list
becomes:
tests.package.test_xen.TestXenAarch64
tests.package.test_xen.TestXenArmv7
[1] 5346824a83
Cc: Vincent Stehlé <vincent.stehle@arm.com>
Reviewed-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
Add a TestXenArmv7 class with its related files in an arm/ subfolder
under test_xen/, to test Xen on 32-bit Arm v7.
We cannot boot with UEFI in this case; we use a custom U-Boot script,
which creates the Xen configuration Devicetree during boot.
Signed-off-by: Vincent Stehlé <vincent.stehle@arm.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
In preparation of adding a test for Xen on 32-bit Arm v7:
- Introduce an architecture-agnostic TestXenBase class where we move
most of the Xen test scenario and bits of the configuration.
- Re-organise the test_xen/ folder with the architecture-agnostic files
under common/ and the 64-bit Arm specific files under aarch64/.
Make the 64-bit Arm TestXen class inherit from the base class and leave
in there only the architecture-specific parts:
- The 64-bit Arm configuration bits.
- The test function, which passes the proper 64-bit Arm simulator
options to the generic test function.
No functional change intended.
Signed-off-by: Vincent Stehlé <vincent.stehle@arm.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
Via the docker-compose runtime test, we can exercise the full suite of
the docker-related packages: docker-compose, of course, but also
docker-engine, which in turn allows exercising containerd. The latter
by defualt uses runc as the container runtime, but can alternatively use
crun.
Extend the docker-compose runtime test with a variant that enables crun.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
The syslinux runtime test (which is in fact a build-only test) ensures
that syslinux does get build at least once a week (via the gitlab-CI
weekly pipeline). Runtime testing would need much more work, though, but
nothing in syslinux is currently runtime tested anyway.
Reported-by: Julien Olivain <ju.o@free.fr>
Signed-off-by: Yann E. MORIN <yann.morin@orange.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
This package provides a reference, functional implementation of a
registry for distributing container images. Formerly the Docker
registry, it now also handles OCI images, so can be used to distribute
all kinds of container related images: classic Docker images, OCI
images, but also helms [0] and the likes, which are also packages as OCI
images in fact.
The upstream name of the package is just "distribution" [1], but this is
a bit too generic and unsearchable; distributions like Debian and Ubuntu
traditionally provides it as the "docker-registry" package, but this is
now misleading as it is not just for Docker anymore. So we decided to
name it "distribution-registry": it contains the upstream name and will
be searchable with the 'registry' part as well.
The built executable is named "registry", which is also a bit too
generic. Again, for the same reasons as above, we name it
"distribution-registry"; this also incidentally makes our golang-package
infra "just work" and do the right thing when it installs the package.
The upstream package provides a wrapper Makefile to build the package,
but it does not do more than our infra already does; the only little nit
is that it allows updating the version file (with git infra et al.). We
can just ignore that and update the version ourselves.
We also provide a minimalist configuration file: it exposes the registry
to localhost only, as a security measure. This file will have to be
overridden (e.g. in a rootfs-overlay) with proper authentication, and
with further locally meaningful setup anyway.
Finally, we also implement a run-time test that validates that we can
push and pull images to/from the registry. Since the registry is a big
go package, it takes time to start, so wait for it a little bit (and yet
a little bit more after it starts logging), and increase the timeouts
for commands that deal with the registry.
For this simple test, the registry is directly exposed with plain HTTP,
not HTTPS, so we must declare it as insecure, so that container tools
can reach it without custom options (e.g. without using skopeo's
--{src,dest}-tls-verify=false options).
We re-use skopeo to talk to the registry. This allows us to grab an
image from the Docker hub registry, and push it to our local registry,
then retrieve it back. Since we want to diffferentiate failures to grab
the image from the Docker Hub [2], from failures to push to our
registry, we do it in two steps: copy from Docker Hub to a local OCI
image, then push that to our registry.
Since both distribution-registry and skopeo are big go-based, statically
linked binaries, running both in the standard virtual machine does not
work: the 256MiB emulated in vexpress are not enough and cause segfaults
when running both at once. The vexpress also only accepts at most
256MiB, so we must use another machine; we decided to use the same as is
used by the podman teest case, which allows at least 1Gib, which is now
enough.
[0] https://helm.sh/
[1] https://github.com/distribution/distribution
[2] in case we reach the 100-pulls-a-day limitation
Signed-off-by: Yann E. MORIN <yann.morin@orange.com>
Cc: Christian Stewart <christian@aperture.us>
Signed-off-by: Julien Olivain <ju.o@free.fr>
... and extend the skopeo runtime test to make use of those.
Signed-off-by: Yann E. MORIN <yann.morin@orange.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
The new test requires a br2-external directory because we compile a
small test program on the host and install it on the target, but it's
not useful to have it in the main Buildroot package tree.
The test program loads and parses a sample HTML document. Taking
inspiration from 'examples/get_title.c' in gumbo-parser, it also
searches for the title of the document just to check that we can do
more than the parsing.
Signed-off-by: Raphaël Mélotte <raphael.melotte@mind.be>
Signed-off-by: Julien Olivain <ju.o@free.fr>
This test verifies that we can run nginx with the modsecurity
directives.
It also checks a very simple rule that blocks requests containing the
keyword "blockme".
Signed-off-by: Raphaël Mélotte <raphael.melotte@mind.be>
[Julien:
- add / at directory end in DEVELOPERS
- sort DEVELOPERS entries alphabetically
- remove unneeded test configs already present in
BASIC_TOOLCHAIN_CONFIG
- sort test config directives alphabetically
]
Signed-off-by: Julien Olivain <ju.o@free.fr>
For release notes since edk2-stable202411, see:
https://github.com/tianocore/edk2/releases/tag/edk2-stable202502https://github.com/tianocore/edk2/releases/tag/edk2-stable202505
Note: the edk2-stable202505 release note mention a security fix.
This fix was already backported in Buildroot in a previous commit.
For that reason, this bump is not marked as securiy, and also
removes the backported patch and its associated _IGNORE_CVES entry.
Those versions include fixes when using gcc 15 (host and target).
This commit also updates the edk2-platforms and edk2-non-osi packages
with the last commit merged at the edk2 release date (2025-05-23).
In the case of edk2-platforms, the commit id used in this update
is few days later, to include a fix for the Arm Sgi575 platform.
Exact commit should have been [1] (to be exactly aligned to the edk2
tag date), but commit [2] is used instead. The package edk2-non-osi is
updated to commit [3].
Also, the SocioNext SynQuacer platforms (including the DeveloperBox)
were removed upstream in edk2-platforms commit [4]. This commit
reflect that change by removing those platforms choices and associated
runtime tests.
This commit has been runtime tested with tests using EDK2 package,
with commands:
support/testing/run-tests \
-d dl -o output_folder \
tests.boot.test_edk2 \
tests.boot.test_grub.TestGrubAArch64EFI \
tests.boot.test_grub.TestGrubi386EFI \
tests.boot.test_grub.TestGrubRiscV64EFI \
tests.boot.test_grub.TestGrubX8664EFI \
tests.fs.test_iso9660.TestIso9660Grub2EFI \
tests.fs.test_iso9660.TestIso9660Grub2Hybrid \
tests.package.test_fwts
It has also been runtime tested (by booting in qemu) with defconfigs
using EDK2 package:
qemu_aarch64_sbsa_defconfig
qemu_riscv64_virt_efi_defconfig
[1] 2bfe3fa99e
[2] 92f7d69e8a
[3] ea2040c2d4
[4] aaba5807f1
Cc: Dick Olsson <hi@senzilla.io>
Cc: Romain Naour <romain.naour@smile.fr>
Cc: Vincent Stehlé <vincent.stehle@arm.com>
Tested-by: Vincent Stehlé <vincent.stehle@arm.com>
Reviewed-by: Romain Naour <romain.naour@smile.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
Back port four upstream fixes for those issues:
- gpsprof could not log or dump GNSS messages; gpsprof is useful
to calbrate a GNSS receiver, especially when building a static
RTK base;
- the JSON blurb reported by gpsd was sometimes broken, causing
decoding errors in clients;
- the detection of 64-bit time on 32-bit systems was borked;
- UDP mode was borked;
The runtime test is extended to test all known transports: TCP, UDP,
and PTY.
Signed-off-by: Yann E. MORIN <yann.morin@orange.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
This test verifies that clinfo correctly detects the rustiCL platform
and llvmpipe as device.
We check:
- exit code,
- Platform name matches rusticl,
- Device name matches llvmpipe.
Signed-off-by: El Mehdi YOUNES <elmehdi.younes@smile.fr>
[Julien:
- rename file to test_mesa3d.py
- add Medhi in DEVELOPERS for this test file
]
Signed-off-by: Julien Olivain <ju.o@free.fr>
For change log, see:
https://gitlab.com/gpsd/gpsd/-/blob/release-3.26.1/NEWS
Support for RTCM104v2 and v3, as well as for ublox, are now
unconditionally built, so drop our corresponding options. No
needfor legacy options: users that had those enabled will
still get them.
Support for Oceanserver has been dropped upstream, so drop our
option and add legacy handling to infrom users they should
complain upstream if they needed it (very unlikely).
Also, a bug introduced in gpsd 3.26 makes the Buildroot gpsd runtime
test failing. See [1]. As a workaround, this commit also changes
this runtime test to emulate a gps over TCP, instead of UDP.
[1] https://gitlab.com/gpsd/gpsd/-/issues/337
Signed-off-by: Yann E. MORIN <yann.morin@orange.com>
[Julien:
- rebased on branch next
- add gpsd version removing OceanServer option in Config.in.legacy
- add link to NEWS in commit log
- switch test_gpsd to use tcp instead of udp
]
Signed-off-by: Julien Olivain <ju.o@free.fr>
The aarch64 virt platform doesn't have any default VGA devices so we
don't need to configure them here.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Julien Olivain <ju.o@free.fr>
pydantic pins an exact version of pydantic-core in it's pyproject.toml
[1]. This was missed during the last pydantic bump where pydantic-core
was erroneously updated to the latest release instead, in commit [2]
and [3].
Modify the pydantic runtime test to catch this error in the future.
[1] https://github.com/pydantic/pydantic/blob/v2.11.4/pyproject.toml#L49
[2] e1c939e426
[3] e42de820cd
Signed-off-by: Marcus Hoffmann <buildroot@bubu1.eu>
[Julien: add reference to commits introducing the issue]
Signed-off-by: Julien Olivain <ju.o@free.fr>
Since Buildroot commit [1], the runtime test emulator infra is setting
the emulated system date to the host date.
While this is desired in general, this behaviour is introducing a
variability in the test execution. Depending if the test is executed
during winter or summer time, the output of the command "date +%Z"
will produce a different output.
This commit fixes the issue by setting a fixed date and time on the
emulated system. The date is fixed to Unix Epoch plus one hour. This
is because Linux cannot set the system date to a value less than the
system uptime. So we cannot set the time back to Unix Epoch with the
command "date -s @0" (this would result to a EINVAL Invalid argument).
Fixes:
https://gitlab.com/buildroot.org/buildroot/-/jobs/9922589073https://gitlab.com/buildroot.org/buildroot/-/jobs/9922589081
[1] cf8641b73e
Signed-off-by: Julien Olivain <ju.o@free.fr>
Add a kernel fragment to enable
CONFIG_ARM64_VA_BITS_48 for the AARCH64 QEMU platform in the
clang compiler -rt runtime test.
Without this setting, this sanitize-based test fails with clang-20
due to insufficient virtual adress space.
"This might potentially affect ARM platforms with 39-bit address space." from [1]
[1] a588cfe37e
Signed-off-by: El Mehdi YOUNES <elmehdi.younes@smile.fr>
Signed-off-by: Bernd Kuhls <bernd@kuhls.net>
Signed-off-by: Julien Olivain <ju.o@free.fr>
The test_xen.py runtime test, introduced in [1] and improved in [2],
is calling a "stty raw" command, just after the emulated machine
login, to avoid double-cooking the consoles. This double-cooking
prevents the test controller to correctly get the command error codes.
Buildroot commit [3] "support/testing: set date in emulated machine"
introduced an invocation of the date command to set time on the
emulated machine, just after the login. The returned error code is also
checked. Since this commit [3], the test_xen runtime test is failing
while attempting to set the date. This is because it is invoked before
the test script executes this "stty raw" command.
The need of executing a command just after the login, and just
before we set the emulated machine date is very limited. It is almost
specific to this test. So, rather than changing the test
infrastructure, this commit simply moves this "stty raw" invocation
from the runtime test script to a custom /etc/profile.d/stty-raw.sh
file on target rootfs overlay, to do this call just at the login.
Fixes:
https://gitlab.com/buildroot.org/buildroot/-/jobs/10000011350
[1] 055f82ebbd
[2] cd0ffd598c
[3] cf8641b73e
Cc: Vincent Stehlé <vincent.stehle@laposte.net>
Tested-by: Vincent Stehlé <vincent.stehle@laposte.net>
Signed-off-by: Julien Olivain <ju.o@free.fr>
Note: this test was not working in Buildroot test infrastructure
before commit [1] was merged, because dieharder has the string "# "
in its output.
[1] 0cad947b96
Signed-off-by: Julien Olivain <ju.o@free.fr>
Signed-off-by: Arnout Vandecappelle <arnout@rnout.be>
While in theory, the fastapi tests finds problems with the pydantic
package, it's not obvious that this test should be run when the pydantic
package is updated.
Add a new test that just covers pydantic.
Signed-off-by: Marcus Hoffmann <buildroot@bubu1.eu>
Signed-off-by: Arnout Vandecappelle <arnout@rnout.be>
Add a runtime test for the 'dust' package to verify that the binary
executes correctly in a minimal buildroot rootfs. The test checks that:
- 'dust --version' runs without error
- 'dust' can analyze a directory structure with files
- The output includes the expected directory names
Signed-off-by: El Mehdi YOUNES <elmehdi.younes@smile.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
Add a runtime test for the 'bat' package to verify that the binary executes
correctly in a minimal Buildroot rootfs.The test cheks that:
- 'bat --version' runs without error
- 'bat' can read and display a text file
- the displayed content matches the expected string
Signed-off-by: El Mehdi YOUNES <elmehdi.younes@smile.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
The test timeout is no longer sufficient, so double it.
Signed-off-by: Raphaël Mélotte <raphael.melotte@mind.be>
Signed-off-by: Julien Olivain <ju.o@free.fr>
Importing urllib3 already allows us to check that for example zlib can
be loaded at runtime.
For good measure, also create the PoolManager object mentioned in the
user guide ([1]), and check that we can normalize a URL like it is
done in urllib3's 'test/test_util.py'.
[1]: https://urllib3.readthedocs.io/en/stable/user-guide.html
Signed-off-by: Raphaël Mélotte <raphael.melotte@mind.be>
Signed-off-by: Julien Olivain <ju.o@free.fr>
Commit [1] "support/testing: improve weston test reliability" moved
out the wait time from the emulator (to run on the test controller).
While doing so, the sleep time which was initially _after_ the
"killall weston" invocation to in stop_weston() was incorrectly
moved before the command invocation. In this state, the test can
succeed on fast host computer running the test. But it will most
likely fail on an average computer.
This commit fixes this issue by moving the sleep time after
the command invocation.
[1] 6561a5d773
Signed-off-by: Julien Olivain <ju.o@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Some EDK II configurations have complex dependencies on several packages
and additional build options; build tests help keeping track of those
more easily.
Factorize some code common to all the build tests into a new
TestEdk2BuildBase class, which defines a base configuration and a method
to assert that binaries do indeed exist after the build.
While at it, add myself in DEVELOPERS.
Signed-off-by: Vincent Stehlé <vincent.stehle@arm.com>
Cc: Dick Olsson <hi@senzilla.io>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Switch the test assert to use 'Europe/Berlin' instead of
'posix/Europe/Berlin'. The tz-name is an internal implementation detail
anyway and the lookup changed to prefer the stated name in /etc/timezone
again, if it matches the configured timezone via /usr/share/zoneinfo.
Ref: 4ffac9232b
Signed-off-by: Marcus Hoffmann <buildroot@bubu1.eu>
Signed-off-by: Julien Olivain <ju.o@free.fr>
Since Buildroot commit [1], the test_gstreamer1 runtime test is
failing to build the tesseract-ocr package. The root cause is that
the test uses the default arm external toolchain, which is the Linaro
ARM 2018.05 based on gcc 7.3.1.
Since [1], tesseract-orc no longer compiles with gcc 7 (it requires at
least gcc 8).
This commit fixes the issue by switching the toolchain to the
bootlin versions (based on gcc 14 at the time of this commit).
Also, changing the compiler version slightly slowed down the video
encoding. This commit also increases the encoding command timeout.
Fixes:
https://gitlab.com/buildroot.org/buildroot/-/jobs/9407846232
[1] f32da8b984
Signed-off-by: Julien Olivain <ju.o@free.fr>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
The busybox date applet accepts the following:
date @1234567
but this confuses the coreutils version which doesn't implicitly set
time. As some tests might need coreutils binaries we should ensure the
emulator login will work with both. Fix this by passing the -s (set)
option to the command.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Julien Olivain <ju.o@free.fr>
In commit ba70f94e62 (subversion makes it so we do not need a separate
tarball to store empty directories and symlinks for the target_skeleton.)
/etc/resolv.conf was made a symlink to ../tmp/resolv.conf. There was no
rationale behind this change, but back at the time (2006!), the only
location that Buildroot guaranteed to be writable was /tmp, and we
needed a writable location to create resolv.conf from a DHCP reply for
example, so the change made sense.
Commits 2f67d9e2fe (preparation for BR2_PACKAGE_BUSYBOX_SKELETON), in
2007, then d9119c0bc5 (move default skeleton to fs/ and drop busybox
skeleton), in 2010, and finally 6c3e3ad419 (New top-level directory:
system), in 2012, propagated that /etc/resolv.conf without any change in
the target of the symlink.
However, since commit d380dde1d2 (skeleton: make /run a proper
directory/filesystem), in 2015, a tmpfs has been mounted on /run, so it
is now guaranteed to be writable as well for SYS-V-based systems. That
has also been the case for openrc since commit a840e485d1
(package/skeleton-init-openrc: new package), in 2019. This has also
always been the case for systemd as well.
resolv.conf is not a temporary file [0]; it is really more of a runtime
data; its location in /tmp is just a historical artefact of a long-wrong
assertion.
Nowadays, distributions either have an actual file as /etc/resolv.conf
(when /etc is writable), or it is a symlink to a location below /run.
Although resolv.conf is used by the C library resolver, a lot of
applications now will directly use it; this is the case for some
container management tools, such as podman (and its helpers, like
netavark), that expect /etc/resolv.conf to either be a file or a symlink
to somewhere in /run, and fail to provide DNS services if that is not
the case.
So, we move the target of /etc/resolv.conf to ../run/resolv.conf.
As a consequence, we can drop the workaround in podman's runtime test.
To be noted: there are tow custom fstab files in Buildroot that do not
explicitly mount a tmpfs on /run:
- one for a systemd runtime test, so /run would also be a tmpfs there
anyway, as systemd does mount one;
- one for the mender_x86_64_efi_defconfig; that's a not a problem:
/ is an ext4 that is mounted r/w (and /tmp is not a tmpfs either, so
the symlink to /tmp would not not have worked if / was not r/w
anyway).
This leaves people with their own fstab, to adapt. Cater to them by
adding a little blurb in the documentation.
Note: the following 43 tests all built and ran successfully:
tests.init
tests.package.test_apache
tests.package.test_docker_compose
tests.package.test_dropbear
tests.package.test_iputils
tests.package.test_podman
tests.package.test_skopeo
tests.package.test_wget
(no other runtime test were attempted)
[0] temporary, as in relatively short-lived.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
passt/pasta is the officially supported network backend, with slirp4netns
only considered a legacy solution. However, some people have experienced
corner cases when using passt/pasta (as it still is a young stack), so we
do not want to do a blanket replace of slirp4netns just yet. Still, we
make passt/pasta the default option.
As a consequence, the existing runtime tests will now test passt/pasta,
so we introduce two new tests for slirp4netns as a network backend.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
Podman is a container manager not unlike Docker, but is daemon-less.
Similarly to docker-engine, quite a few kernel config options are
required; as they are very similar in goals and features, the options
from docker-engine have been duplicated for podman. As that was not
enough, a few additional options have been added after trial-and-error
testing (they are not explicitly listed in the documentation).
The documentation [0] states that seccomp can be disabled (i.e. not
enabled). However, without seccomp support, starting containers requires
--security-opt=seccomp=unconfined to be specified; it does not look
trivial to make that the default, though [1]. Furthermore, containers
are about security, so disabling a security measure does not sound too
good. So we make seccomp support mandatory.
Podman needs netavark as a the network backend (it missing is a hard
error at runtime). It is supposed to also require aardvark-dns, an
authoritative DNS resolver, but it missing does not look to adversely
affect networking, so it is not used (as not even packaged in Buildroot
yet).
Podman can run either as the root user, or it can run rootless, i.e. by
a non-root user, which requires a bit of setup (see below, in the
runtime test) and a few other dependencies: slirp4netns [2] (to provide
network connectivity in userland), and support for subordinate UIDs and
GIDs with the shadow library. Rootless mode is one of the main selling
point for podman, so we decided that this would not be configurable in
Buildroot.
Similar to Docker, podman can inject a minimalist init as PID1 in
containers, and like Docker, this is optional; podman however by default
uses catatonit as such an init [3]. As for Docker-engine, we offer a
choice of which init package to use to provide podman-init.
Podman requires at least three config files; they can be either per-user
or system-wide:
- containers.conf [4]: defines various settings for the container
runtimes;
- policy.json [5]: defines what signature to accept to validate
images; without one such file, podman just refuses to pull images;
- registries.conf [6]: defines where to pull images from; without it,
podman does not know how to pull un-qualified images (i.e. images
where the registry is not specified in the path, and which Docker
would fetch from the Docker Hub, e.g. "busybox:latest").
For those three files, we provide a very minimal default that (in the
same order as above):
- uses the slirp4netns network backend for rootless operation (the
default in podman is to use pasta [1], so we need to explicitly
configure it to use slirp4netns);
- allows pulling images which signature can't be verified;
- pulls unqualified images from the Docker Hub, as is traditional.
Providing actual files is going to be use-case dependent, and interested
parties will have to provide their own config files, e.g. in a rootfs
overlay.
Finally, we add a runtime test for podman. Podman is a huge binary, and
may call other huge binaries (netavark...); this can be quite slow in
the emulated machine (even when running on a very fast host machine), so
we use a huge timeout for all commands involving podman, even those that
exit the containers, as that may need to tear down podman setup.
The default kernel used in runtime tests is missing a lot of features,
so we need to build our own; we use the same version as the bundled
kernel. We can't use cpio either, because we need a filesystem that can
be used as a lower and upper of overlayfs, which is not possible with
the filesystem the cpio is extracted into; ext2 fits the bill, so we use
that. We need a bit of space to store images and stuff, so let's be
generous and allocate 256M.
To test rootless operation, we need a non-root user that has some
special setup [7]; it is easier to run the commands from the infra
rather than carry a user-definition table and a rootfs overlay. We need
that user to have the same prompts (main and continuation) so that the
REPLWrapper still detects those; it has the unfortunate side effect that
it is not immediately obvious whether a command was run as root or not,
and one has to look back up in the run-log to see whether there was a
transition to another user earlier.
Still for rootless containers, podman/netavark expect /etc/resolv.conf
to be either a plain file, or a symlink that points either deeper in
/etc or anywhere in /run; if resolv.conf resolves to any other location,
DNS in rootless containers does not work. This is reasonable, and is
what already happens on a systemd-based system (and thus all major
distributions nowadays. However, in Buildroot, we put the actual file in
/tmp; this is historical, and dates back to the days where Buildroot did
not have a guaranteed-writable /run. So, we work around this limitation
in the test (for now).
The official busybox image on the Docker Hub supports a lot of
architectures, of which armv7 which we use for this runtime test.
Finding a small image that also supports armv7 on other registries was
a bit of a challenge; we eventually found one busybox image on quay.io,
but it is not an official busybox image; still, it fits the bill, so we
use it.
There is no runtime test with systemd, as this requires quite some
additional setup that does not look very trivial to do; when it detects
it is running under systemd in rootless mode, podman expects that a full
user session exists, or it whines about it every time it is started,
reverting to non-systemd behaviour; getting a full user session does not
look to be that trivial (PAM?), so this would not exercise the actual
integration with systemd, so the test would not be meaningful, so it is
not provided. This is left as an exercise to an interested party to
extend the tests.
PS: Hat-tip to Raphael, who provided some pointers and hints on this
change, especially for rootless mode. Thanks! 👍
[0] https://podman.io/docs/installation#get-source-code
[1] it looks like we can provide a custom seccomp profile, by specifying
seccomp_profile="PATH" in containers.conf; that would still require
seccomp support to use that file, though, so that does not change the
outcome.
[2] it is possible to use another backend, but it is not packaged in
in Buildroot yet: https://passt.top/passt/about/#pasta-pack-a-subtle-tap-abstraction
[3] podman expects a 'catatonit' helper in /usr/libexec/podman, so even
if tini would be usable instead, it would not feel right to use it to
impersonate catatonit. So let's assume that only catatonit is supported.
[4] https://github.com/containers/common/blob/main/docs/containers.conf.5.md
[5] https://github.com/containers/image/blob/main/docs/containers-policy.json.5.md
[6] https://github.com/containers/image/blob/main/docs/containers-registries.conf.5.md
[7] https://github.com/containers/podman/blob/main/docs/tutorials/rootless_tutorial.md
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Raphael Pavlidis <raphael.pavlidis@gmail.com>
Cc: Christian Stewart <christian@aperture.us>
Cc: Julien Olivain <ju.o@free.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
Skopeo can be used to manipulate container images, and so can be useful
on target devices that are going to be used as a host to run containers.
So, add a target variant for skopeo.
... with a runtime test while at it.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Christian Stewart <christian@aperture.us>
Cc: Yann E. MORIN <yann.morin@orange.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
When time comes to check certificates, the date and time in the emulated
machine should be close enough to the actual values, so that certificate
validity can be checked.
Some Qemu machines have an RTC (e.g. arm vexpress-a9 has a pl031), and
the kernel needs a driver for those RTC. It is not guaranteed that the
machine used for a test meets those two conditions; in such a case, the
time in the machine starts way back in the past (1970-01-01T00:00:00Z on
sysv, or the release date of systemd). This is the case with the default
kernel, so such tests do not have the proper time.
Set the date to the date of the host system. This is going to be accurate
to the second, which is, by far, enough for our purpose.
To avoid having to consider what combination of emulated machine and kernel
configuration are being used, we always set the date, as this is a
generic step that should be done by the infra (like login in as root is).
The Emulator() class doesn't inherit from unittest.TestCase, so we can't
call any of the usual self.assertXXX() methods; instead, we just raise
a standard exception, like is done a few lines above to detect the login
prompt.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Ricardo Martincoski <ricardo.martincoski@datacom.com.br>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
