* fix: make assert test for netaddr actually return a boolean
The netaddr test returns a string when the netaddr is installed. This makes
Ansible 2.20 angry. Here's a fix to make sure the true case also returns a
boolean instead of a string.
* fix: more fixes for non-boolean conditions
The `cloud_provider` assertion change is a little more involved. The only two
allowed values are "" and "external". Let's just always check the assertion
instead of skipping it when it's the default value, which is "".
All the other changes should be fairly obvious.
This commit updates documentation across several files to correct broken
file paths and ensure consistency in Ansible command examples
- Standardized inventory naming to `inventory.ini` for sample inventories.
- Fixed `group_vars` paths to reflect the actual directory structure
(e.g., `group_vars/all/all.yml` instead of `group_vars/all.yml`).
- Corrected the `k8s-cluster.yml` filename in guides.
- Clarified the location for `kubectl_localhost` and `kubeconfig_localhost` settings.
- Replaced non-existent placeholders (like `inventory/single.cfg`) with
valid repository paths.
- Fixed typos in directory names (e.g., `myclsuter` -> `mycluster`).
Add kube_vip_metrics_enabled and kube_vip_metrics_port; wire
prometheus_server and container port in the static pod manifest.
Default metrics port to 2112 to match kube-vip upstream. Document
and sample inventory updated.
Fedora 39 (EOL 2024-11-26), Fedora 40 (EOL 2025-05-13), and Fedora 41
(EOL 2025-12-15) are all out of support and no longer receive security
updates. This bumps the kubevirt CI test matrix to currently supported
releases, distributing tests across Fedora 42 (supported until
2026-05-13) and Fedora 43 (supported until 2026-12-09) to keep
multi-version coverage.
- Add fedora-43 image entry to the kubevirt image-builder
- Rename fedora39-* tests to fedora43-* (4 tests)
- Rename fedora40-* tests to fedora43-* (2 tests)
- Rename fedora41-* tests to fedora42-* (4 tests)
- Update .gitlab-ci/kubevirt.yml testcase names
- Regenerate docs/developers/ci.md
All 10 existing Fedora test configurations (kube-router, calico-selinux,
calico-swap-selinux, crio with/without SELinux enforcing, docker-calico,
flannel-crio-collection-scale) are preserved.
Signed-off-by: Kay Yan <kay.yan@daocloud.io>
* image-builder: run validation locally with BuildKit
* upadte checksum
* set BuildKit no-process-sandbox flag for CI
* run BuildKit with rootless-safe daemon flags in CI
* update
* updating root logic
* fix ci failure for can't enable NoProcessSandbox
* switch to stable url
* add --oci-worker-no-process-sandbox ci flag in root path
* add more support
* add validate-docker and validate-single docker
Cilium < 1.20 unconditionally registers a field indexer for TLSRoute
v1alpha2 when the Gateway API controller is enabled, but Gateway API
>= 1.5.0 ships TLSRoute v1alpha2 with served=false in the standard
channel. The result is cilium-operator CrashLoopBackOff with:
no matches for kind "TLSRoute" in version "gateway.networking.k8s.io/v1alpha2"
The fix landed in Cilium 1.20 only and will not be backported.
Add a preflight assert that triggers only when all of the following
hold: cilium_gateway_api_enabled, gateway_api_enabled, cilium_version
< 1.20.0, gateway_api_version >= 1.5.0, and gateway_api_channel ==
"standard". Users hit by this combo get a clear error and two
workarounds (pin gateway_api_version to 1.4.1, or switch
gateway_api_channel to "experimental") instead of debugging a crash
loop after the fact.
Signed-off-by: Kay Yan <kay.yan@daocloud.io>
The documented example still uses the removed --required-author flag and
misses --repo-path, which breaks with current release-notes binaries.
Update it to use the generate subcommand and point at the local checkout.
Signed-off-by: Kay Yan <kay.yan@daocloud.io>
kata containers added create_container_timeout parameter.
If this parameter is missing, the new version of Kata cannot create containers properly and will encounter an error: failed to create shim task: context deadline exceeded
