Support number of workers for the virtual cluster

[SchSeba]

also add a make target to redeploy the containers on the cluster for an easy development cycle

Signed-off-by: Sebastian Sch [email protected]

[github-actions]

Thanks for your PR,
To run vendors CIs use one of:

• /test-all: To run all tests for all vendors.
• /test-e2e-all: To run all E2E tests for all vendors.
• /test-e2e-nvidia-all: To run all E2E tests for NVIDIA vendor.

To skip the vendors CIs use one of:

• /skip-all: To skip all tests for all vendors.
• /skip-e2e-all: To skip all E2E tests for all vendors.
• /skip-e2e-nvidia-all: To skip all E2E tests for NVIDIA vendor.
  Best regards.

[bn222]

This should not be part of upstream

[SchSeba]

agree removing from the Makefile

[bn222]

/lgtm

[adrianchiris]

lets discuss this one in a meeting.

my thoughts:
tests should be agnostic as possible to the k8s distro (upstream, ocp)
any ocp specific prep scripts or ocp specific tests should be out of tree in a different repo.
GH Actions can easily clone more than one repo.

[SchSeba]

@adrianchiris I agree that is why we have a separate repo for OCP specific tests using the sriov operator under the cnf-feature-deploy repo

https://github.com/openshift-kni/cnf-features-deploy/blob/master/cnf-tests/testsuites/e2esuite/dpdk/dpdk.go
https://github.com/openshift-kni/cnf-features-deploy/blob/master/cnf-tests/testsuites/e2esuite/sctp/sctp_sriov.go
https://github.com/openshift-kni/cnf-features-deploy/blob/master/cnf-tests/testsuites/e2esuite/vrf/vrf_sriov.go

and I may in the future clone that repo as part of CI system here, but the script to deploy the cluster (can be vanilla ocp k3s or others) with the PR build operator should be in the operator repo if you ask me.

that is also important so all the maintainers in this repo have full control on the CI lanes like versions and runners

[bn222]

To me, it makes more sense to have as few as possible references to non-vanille k8s in this repo. Again, it's not black and white. I understand the need to trigger a job that is testing a vendor-specific release. That's ok imo as long as there is a single entry point to kick off all kinds of vendor-specific testing. In short, I'm in line with @adrianchiris

Regarding the actual code, I never liked the fact that we had if openshift in upstream. Better to abstract those things away too and call a generic thing instead that can be dynamically replaced by an instance of something specific downstream.

[bn222]

To add some detail to my previous comment, it's important to realize that there are two parts: "vendor specific testing" and "code related stuff". For this PR, let's focus on the first part only.

[github-actions]

Thanks for your PR,
To run vendors CIs use one of:

• /test-all: To run all tests for all vendors.
• /test-e2e-all: To run all E2E tests for all vendors.
• /test-e2e-nvidia-all: To run all E2E tests for NVIDIA vendor.

To skip the vendors CIs use one of:

• /skip-all: To skip all tests for all vendors.
• /skip-e2e-all: To skip all E2E tests for all vendors.
• /skip-e2e-nvidia-all: To skip all E2E tests for NVIDIA vendor.
  Best regards.

[coveralls]

Pull Request Test Coverage Report for Build 6661551394

• 0 of 0 changed or added relevant lines in 0 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.09%) to 25.71%

---

Totals
Change from base Build 6639207535:	0.09%
Covered Lines:	2328
Relevant Lines:	9055

---

💛 - Coveralls

[zeeke]

LGTM