Now that a new kind version is released supporting 1.36,
we can use it for our CI testing.
This commit:
- Bumps the kind version and its images to support the 1.36
- Ensure all the API calls are done with client-go 0.36
(and dependent k8s deps), to allow future 1.37 release.
Signed-off-by: Jean-Philippe Evrard <open-source@a.spamming.party>
We have historically decided to support AND TEST 3 versions
of kubernetes. For that, we lagged behind in terms of client versions.
Lagging behind one version allowed us, thanks to kubernetes client version skew,
to support 3 versions at relatively low cost.
However, maintaining 3 versions has always been a pain.
For that, we pinned old versions of kubernetes clients/api and test tooling.
In december 2025 community meeting, we decided to only maintain two
versions: the most current one, and the previous one.
This allows us to bump more regularly and be more secure.
We can also more easily use dependabot this way.
Signed-off-by: Jean-Philippe Evrard <open-source@a.spamming.party>
Now that a new kind version is released supporting 1.35,
we can use it for our CI testing.
This commit:
- Bumps the kind version and its images to support the 1.35
- Ensure all the API calls are done with client-go (and other k8s deps)
supporting a maximum ranges of versions (0.34)
- Aligns Kured go version with k8s 0.34 go version.
Signed-off-by: Jean-Philippe Evrard <open-source@a.spamming.party>
Now that a new kind version is released supporting 1.34,
we can use it for our CI testing.
This commit:
- Bumps the kind version and its images to support the 1.34
- Ensure all the API calls are done with client-go (and other k8s deps)
supporting a maximum ranges of versions (0.33)
- Aligns Kured go version with k8s 0.33 go version.
Signed-off-by: Jean-Philippe Evrard <open-source@a.spamming.party>
Now that 1.33 kind is released, we can use it to bump to k8s
0.32, which should work for the 3 releases (1.31,1.32,1.33).
This work guarantees it with the help of our CI.
Signed-off-by: Jean-Philippe Evrard <open-source@a.spamming.party>
* Move to stable kind cluster filenames
Without this, we have to rename files at every version.
This is really unnecessary, we should only change the files
and be done with it.
This is a problem, as if we move to programmatic test running,
the tests would need to be mutatated at every k8s version.
With this model, we know that only the kind-cluster files
need to be modified for the tests to ba automatically
adapted.
Signed-off-by: Jean-Philippe Evrard <open-source@a.spamming.party>
* Create e2e from go tests interface
Without this, e2e tests need tons of manual work to
test locally, and the results are not easily exposed.
People are less likely to use the e2e tests if they
are tough to use outside the CI.
This commit makes it easier to run tests locally,
and ensures the CI is closer to the Makefile.
At the same time, this removes debt in the github
worfklows: By switching to newer versions of kind,
we can remove the very old workaround for the
failed to attach pid 1.
Signed-off-by: Jean-Philippe Evrard <open-source@a.spamming.party>
* Add node stays as cordonned test
Without this, impossible to prove that the node stays as cordonned
after a reboot by kured.
This refactor also adds the test in the CI, and makes sure the
CI is a bit simpler, by using matrix more extensively.
Signed-off-by: Jean-Philippe Evrard <open-source@a.spamming.party>
* Use hack dir instead of .tmp
This is more idiomatic.
Signed-off-by: Jean-Philippe Evrard <open-source@a.spamming.party>
---------
Signed-off-by: Jean-Philippe Evrard <open-source@a.spamming.party>