Bump dependencies

[nunnatsa]

The primary goal of this PR is to bump the dependencies. In attempting to bump the dependences @nunnatsa discovered a conflict between kubevirt/client-go and updating the core k8s depenedences. Upon further inspection, it was determined that we only needed the kubevirt/client-go package for our e2e tests, and not the capk controllers themselfs.

So, in order to update the core k8s dependences for the capk controllers without entering into dependency hell, Nahshon separated the dependences for the e2e tests and capk controllers. This removed us from having to have kubevirt/client-go in our primary go.mod file.

In all, his changes are...

• Separate dependencies for the controllers and e2e into separate go.mod files
• Update the code to work with the new dependencies. (that meant moving the validation webhook logic and passing around the context object)
• Removing Passt support (the e2e test is failing and we don't intend to support passt binding anymore)

Bump dependencies and separate e2e test and controller dependencies

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: nunnatsa

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [nunnatsa]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[nunnatsa]

/ok-to-test

[coveralls]

Pull Request Test Coverage Report for Build 6584385022

• 35 of 83 (42.17%) changed or added relevant lines in 6 files are covered.
• 1 unchanged line in 1 file lost coverage.
• Overall coverage increased (+12.3%) to 61.992%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
controllers/kubevirtmachine_controller.go	5	16	31.25%
controllers/kubevirtcluster_controller.go	1	15	6.67%
pkg/webhookhandler/validator.go	26	49	53.06%

Files with Coverage Reduction	New Missed Lines	%
controllers/kubevirtcluster_controller.go	1	45.7%

Totals
Change from base Build 6232901575:	12.3%
Covered Lines:	946
Relevant Lines:	1526

---

💛 - Coveralls

[agradouski]

this PR does a bit more than just bumping dependencies.
perhaps it's worth splitting it to several focused PRs @nunnatsa, to simplify the review?
or, please add more comments about various updates you're making to the logic.
thanks

[nunnatsa]

this PR does a bit more than just bumping dependencies. perhaps it's worth splitting it to several focused PRs @nunnatsa, to simplify the review? or, please add more comments about various updates you're making to the logic. thanks

@agradouski - there is no logic change. All the code changes are required by API changes in the bumped modules. I don't know how to split it to smaller commits, if I want any commit to be compile-able.

[davidvossel]

i don't think we have a ubuntu image for v1.27 yet. The highest image i see is quay.io/capk/ubuntu-2004-container-disk:v1.26.0

[davidvossel]

We don't need to do this for this PR. But if all we care about is immutability (which appears to be the case) we can enforce this on the CRD now with CEL, and avoid a webhook entirely.

[davidvossel]

this PR does a bit more than just bumping dependencies. perhaps it's worth splitting it to several focused PRs @nunnatsa, to simplify the review? or, please add more comments about various updates you're making to the logic. thanks
@agradouski - there is no logic change. All the code changes are required by API changes in the bumped modules. I don't know how to split it to smaller commits, if I want any commit to be compile-able.

Yeah, there are a few things going on here. Here's some further context.

The primary goal of this PR is to bump the dependencies. In attempting to bump the dependences @nunnatsa discovered a conflict between kubevirt/client-go and updating the core k8s depenedences. Upon further inspection, it was determined that we only needed the kubevirt/client-go package for our e2e tests, and not the capk controllers themselfs.

So, in order to update the core k8s dependences for the capk controllers without entering into dependency hell, Nahshon separated the dependences for the e2e tests and capk controllers. This removed us from having to have kubevirt/client-go in our primary go.mod file.

In all, his changes are...

1. Separate dependencies for the controllers and e2e into separate go.mod files
2. Update the code to work with the new dependencies. (that meant moving the validation webhook logic and passing around the context object)
3. Removing Passt support (the e2e test is failing and we don't intend to support passt binding anymore)

[davidvossel]

/lgtm

I updated the description of the PR to better reflect the work that is being done here. This PR is important because it gives us the ability more easily bump dependencies in the future for the controllers while not conflicting with some of the dependencies required for our e2e tests.