linux/meta-virtualization.git/tests, branch master-next

tests: add lxc runtime tests with download-template regression

2026-06-14T03:57:14+00:00

The "lxc was tested" claim during the recent runtime testing sweep was actually transitive — incus runs against LXC libraries under the hood, so incus passing was treated as evidence that LXC itself worked. That inference was wrong: incus uses its own go bindings into liblxc rather than the lxc-* command-line tools, and the breakage Ferry Toth reported on 2026-06-13 sat entirely in templates/lxc-download.in (a script invoked by lxc-create, never reached through incus). The bug would not have been caught by any existing test in the layer. Add tests/test_lxc_runtime.py to close the gap. The suite boots container-image-host with CONTAINER_IMAGE_HOST_EXTRA_INSTALL=lxc, then runs three groups of checks against the live guest: TestLxcInstalled — sanity that lxc-create, lxc-start and lxc --version work at all. Catches packaging and PATH-level regressions. TestLxcDownloadTemplate — explicit regression for the templates-actually-create-DOWNLOAD_TEMP-directory.patch failure mode. Runs `lxc-create --template download` and asserts the broken early-mktemp error string ("mktemp: failed to create file via template '-d…") does not appear in the output. We deliberately do not require the download itself to succeed — the bug fires before any HTTP request, so the test stays meaningful on air-gapped CI where the actual fetch would fail for unrelated reasons. TestLxcContainerLifecycle (@pytest.mark.network) — full end-to-end: create from images.linuxcontainers.org, start, attach, stop, destroy. Marked @network so offline runners deselect it cleanly. The regression test above is the primary guard; this is depth. Also register the lxc marker in pytest.ini so collection doesn't warn. The test conventions (pexpect-driven runqemu boot, marker-delimited command runner, TERM=dumb to suppress shell integration escape sequences) match test_incus_runtime.py and test_xen_runtime.py so the three suites read consistently. Signed-off-by: Bruce Ashfield

tests: add requirements.txt

2026-06-12T18:48:59+00:00

Add requirements.txt to allow versions to be pinned. Signed-off-by: Tim Orling Signed-off-by: Bruce Ashfield

tests: add containerd and CRI-O recipe and runtime tests

2026-06-12T02:58:55+00:00

35 tests across three tiers: Tier 1 (static): containerd recipe checks (virtual/containerd provider, systemd service, CNI networking, ctr/shim install), profile config, packagegroup, nerdctl and cri-tools recipe existence, CRI-O recipe checks (seccomp requirement, systemd service, CNI/conmon/runtime deps). Tier 2 (build): containerd, nerdctl, and container-image-host with containerd profile build verification. Tier 3 (boot): containerd service status, ctr/nerdctl availability and version, namespace listing, runtime availability, CNI plugins, nerdctl pull/run/images cycle with busybox. The boot tests build container-image-host with CONTAINER_PROFILE=containerd automatically. CRI-O boot tests are not included — CRI-O requires a kubelet to drive it, so runtime testing is covered by k3s tests when k3s is configured to use CRI-O as its CRI. Signed-off-by: Bruce Ashfield

tests: add libvirt recipe and runtime tests

2026-06-12T02:58:55+00:00

46 tests across three tiers: Tier 1 (static, no build): Recipe structure, PACKAGECONFIG options, hook_support.py python3 compatibility (text mode, raw string regex, syntax validation), service files, patch files, kvm-image-minimal recipe checks. Tier 2 (build): libvirt and kvm-image-minimal build verification. Tier 3 (boot): libvirtd service status, virtlockd socket, virsh connectivity via monolithic daemon socket, capabilities, nodeinfo, domain listing, default network, hook script installation and permissions, qemu user and libvirt group existence. The boot tests use the explicit monolithic daemon socket path (qemu+unix:///system?socket=/var/run/libvirt/libvirt-sock) because libvirt v12 defaults to modular daemons (virtqemud) but the kvm-image-minimal recipe runs the monolithic libvirtd. Signed-off-by: Bruce Ashfield

tests: remove redundant DISTRO_FEATURES appends from build fixtures

2026-06-12T02:58:55+00:00

The -R conf files used by test build fixtures appended DISTRO_FEATURES values that are already provided by meta-virt-host.conf (virtualization, vcontainer, systemd). BitBake tracks append operations in task hashes, so a duplicate append changes the hash and invalidates sstate for every recipe that depends on DISTRO_FEATURES — causing full rebuilds even when the image was just built. Only append features genuinely needed by the test profile and not already present in the base config: - k3s: append only "k3s" (virtualization already in base) - xen: append only "xen vxn" (virtualization systemd vcontainer in base) - incus: no append needed (CONTAINER_PROFILE alone is sufficient) Signed-off-by: Bruce Ashfield

k3s: fix pause image reference for containerd v2

2026-06-12T02:58:55+00:00

K3s's default pause image reference "rancher/mirrored-pause:3.6" is a short name that containerd v1 auto-expanded to the fully qualified "docker.io/rancher/mirrored-pause:3.6". The k3s v1.36 uprev moved to containerd v2, which uses a new CRI plugin (io.containerd.cri.v1) with a pinned_images config instead of the old sandbox_image. Containerd v2 no longer auto-expands short names when looking up pinned images, causing every pod sandbox creation to fail with "image not found" even though the image is present in the containerd image store under its fully qualified name. Pass --pause-image docker.io/rancher/mirrored-pause:3.6 in both the server and agent systemd service files to use the fully qualified reference that containerd v2 can resolve. Signed-off-by: Bruce Ashfield

tests: make k3s, incus, and xen tests self-contained

2026-06-12T02:58:55+00:00

These boot tests previously required manual local.conf changes to set CONTAINER_PROFILE, DISTRO_FEATURES, and other variables before the image could be built and tested. This meant complete test coverage required editing local.conf between runs. Add build fixtures that use bitbake -R with a temporary conf file to inject the required variables automatically: - k3s: builds container-image-host with CONTAINER_PROFILE=k3s-host - incus: builds container-image-host with CONTAINER_PROFILE=incus - xen: builds xen-image-minimal with xen/vxn DISTRO_FEATURES Each test module now builds its own image before booting, using the same run_bitbake + extra_vars pattern established in test_container_cross_install.py. Signed-off-by: Bruce Ashfield

tests: fix image ID parsing and 9p assertion patterns

2026-06-12T02:58:55+00:00

test_vdkr_registry.py: Docker's images output changed from separate REPOSITORY TAG IMAGE_ID columns to combined IMAGE ID DISK_USAGE columns. The parser grabbed the disk usage (8.45MB) as the image ID. Handle both old and new formats. test_container_registry_script.py: The secure registry tests checked for literal 'virtfs' and 'cashare' strings in vrunner.sh, but the 9p share setup was refactored to use hv_build_9p_opts(). Update assertions to match the current abstraction. Signed-off-by: Bruce Ashfield

tests: parametrize boot tests over container profiles for full coverage

2026-06-12T02:58:55+00:00

The boot tests (TestBundledContainersBoot, TestCustomServiceFileBoot) previously depended on whatever CONTAINER_PROFILE was in local.conf, skipping docker checks when podman was configured and vice versa. Getting complete coverage required manually editing local.conf and running the tests twice. Add --container-profiles pytest option (default: docker,podman) and a profiled_session fixture that builds container-image-host with each profile via a temporary bitbake -R conf file, boots it, and runs all checks. Both docker and podman get a full build-boot-verify cycle in a single test run with no local.conf changes. Also fix: - run_bitbake() gains extra_vars parameter for bitbake -R overrides - Strip OSC 3008 shell integration escape sequences from run_command output (BusyBox shell emits these, corrupting file paths used in subsequent commands) - Use known file paths for service file content checks instead of extracting paths from ls output (avoids escape sequence corruption) - Add test_bundle_class_unpack_enabled to catch do_unpack[noexec] regression Signed-off-by: Bruce Ashfield

tests: fix container-cross-install bitbake invocation and busybox compat

2026-06-12T02:58:55+00:00

Three issues in test_container_cross_install.py: - run_bitbake() called bitbake directly, requiring the OE build environment to be pre-sourced. test_multilayer_oci.py wraps bitbake in 'bash -c source oe-init-build-env && bitbake' which is self- contained. Adopt the same pattern so tests work from any shell. - test_vdkr_initramfs_create and test_vpdmn_initramfs_create built the initramfs recipes in the main config, but these recipes require the vruntime multiconfig distro. Use mc:vruntime-x86-64: prefix. - test_systemd_services_directory_exists used 'head -5' which BusyBox does not support (requires 'head -n 5'). The container-image-host image uses BusyBox when built without coreutils. Also fix conftest.py VdkrRunner and VpdmnRunner ensure_memres() to default no_registry=True, so pulled images use short names (alpine:latest) rather than registry-prefixed names from the baked-in VDKR_DEFAULT_REGISTRY config. Signed-off-by: Bruce Ashfield