The dry-run fails now because we are trying to install a CRD and a custom resource (of that CRD type) in the same step.
Signed-off-by: Matt Moyer <moyerm@vmware.com>
- Also correct the webhook url setting in prepare-for-integration-tests.sh
- Change the bcrypt count to 10, because 16 is way too slow on old laptops
Signed-off-by: Ryan Richard <richardry@vmware.com>
I also started updating the script to deploy the test-webhook instead of
doing TMC stuff. I think the script should live in this repo so that
Pinniped contributors only need to worry about one repo for running
integration tests.
There are a bunch of TODOs in the script, but I figured this was a good
checkpoint. The script successfully runs on my machine and sets up the
test-webhook and pinniped on a local kind cluster. The integration tests
are failing because of some issue with pinniped talking to the test-webhook,
but this is step in the right direction.
Signed-off-by: Andrew Keesler <akeesler@vmware.com>
This script was basically an alias for `./hack/module.sh unittest`. We even
tell people to run the unit tests via module.sh in our contributing doc.
Let's ditch it - the best line of (shell code) is the one you don't write.
An analagous change was made in CI to use module.sh in place of test-unit.sh.
Signed-off-by: Andrew Keesler <akeesler@vmware.com>
- Upgrade from `1.19.0-rc.0` to the newly-release `1.19.0`.
- Downgrade from `1.18.6` to `1.18.2` to match some downstream consumers.
Signed-off-by: Matt Moyer <moyerm@vmware.com>
- Indicate the success or failure of the cluster signing key strategy
- Also introduce the concept of "capabilities" of an integration test
cluster to allow the integration tests to be run against clusters
that do or don't allow the borrowing of the cluster signing key
- Tests that are not expected to pass on clusters that lack the
borrowing of the signing key capability are now ignored by
calling the new library.SkipUnlessClusterHasCapability test helper
- Rename library.Getenv to library.GetEnv
- Add copyrights where they were missing
- Makes it easier to guess/remember what are the legal arguments
- Also update the output a little to make it easier to tell
when the command has succeeded
- And run tests using `-count 1` because cached test results are not
very trustworthy
We don't want people to run codegen.sh directly, because it is meant
to be driven by hack/module.sh. To discourage this behavior, we will hide
codegen.sh away in hack/lib. I don't think this is actually what the
hack/lib directory is for, though...meh.
Signed-off-by: Andrew Keesler <akeesler@vmware.com>
Runs code generation on a per-module basis. If `CONTAINED` is not set
the code generation is run in a container.
Mount point in docker is randomzied to simulate Concourse.
Introduce K8S_PKG_VERSION to make room to build different versions
eventually.
find(1) seems to look at directory entries in the order in which they exist
in the directory fs entry. Let's sort these so that we get the same results
regardless of the order of the directory entries.
Signed-off-by: Andrew Keesler <akeesler@vmware.com>
Wow fun times with symlinks. We *think* this script should work in CI
now...but we'll see.
Previously we were seeing a false positive where even though the generated
code was out of date, the CI step did not report failure.
Signed-off-by: Andrew Keesler <akeesler@vmware.com>
Our unit test command is going to get slighly more complex in a future revision. This should let us avoid having to sync the CI pipeline definition so many times.
Signed-off-by: Matt Moyer <moyerm@vmware.com>
