CNV-40881: kubevirt, e2e, add test for advanced multinet

[qinqon]

What this PR does / why we need it:
This change add a test to start a kubevirt hosted cluster with just secondary interface as primary network and start a companion dnsmasq pod attached to that network that acts as dhcp server, gateway and masquerade. The test only has one worker so this works as expected.

Which issue(s) this PR fixes (optional, use fixes #<issue_number>(, fixes #<issue_number>, ...) format, where issue_number might be a GitHub issue, or a Jira story:
Fixes #

Checklist

• Subject and description added to both, commit and PR.
• Relevant issues have been referenced.
• This change includes docs.
• This change includes unit tests.

[netlify]

✅ Deploy Preview for hypershift-docs ready!

Name	Link
🔨 Latest commit	7e8e3f3
🔍 Latest deploy log	https://app.netlify.com/sites/hypershift-docs/deploys/6645c05f9d7fe7000856d60c
😎 Deploy Preview	https://deploy-preview-3902--hypershift-docs.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify site configuration.

[qinqon]

/area test

[qinqon]

/area e2e

[openshift-ci]

@qinqon: The label(s) area/e2e cannot be applied, because the repository doesn't have them.

In response to this:

/area e2e

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[openshift-ci]

@qinqon: The label(s) area/test cannot be applied, because the repository doesn't have them.

In response to this:

/area test

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[qinqon]

error performing canary route check	{"error": "error sending canary HTTP Request: Timeout: Get \"https://canary-openshift-ingress-canary.apps.example-sr92d.apps.cluster-1713337633-ellorent-test.devcluster.openshift.com\": context deadline exceeded (Client.Timeout exceeded while awaiting headers)"}

[openshift-ci]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: qinqon
Once this PR has been reviewed and has the lgtm label, please assign csrwng for approval. For more information see the Kubernetes Code Review Process.

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

Needs approval from an approver in each of these files:

• OWNERS

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

[davidvossel]

the approach makes sense to me. i left one minor comment

[openshift-ci-robot]

@qinqon: This pull request references CNV-40881 which is a valid jira issue.

Warning: The referenced jira issue has an invalid target version for the target branch this PR targets: expected the story to target the "4.16.0" version, but no target version was set.

In response to this:

What this PR does / why we need it:
This change add a test to start a kubevirt hosted cluster with just secondary interface as primary network and start a companion dnsmasq pod attached to that network that acts as dhcp server, gateway and masquerade. The test only has one worker so this works as expected.

Which issue(s) this PR fixes (optional, use fixes #<issue_number>(, fixes #<issue_number>, ...) format, where issue_number might be a GitHub issue, or a Jira story:
Fixes #

Checklist

• Subject and description added to both, commit and PR.
• Relevant issues have been referenced.
• This change includes docs.
• This change includes unit tests.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the openshift-eng/jira-lifecycle-plugin repository.

[qinqon]

/jira refresh

[openshift-ci-robot]

@qinqon: This pull request references CNV-40881 which is a valid jira issue.

Warning: The referenced jira issue has an invalid target version for the target branch this PR targets: expected the story to target either version "4.16." or "openshift-4.16.", but it targets "cnv-4.16" instead.

In response to this:

/jira refresh

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the openshift-eng/jira-lifecycle-plugin repository.

[openshift-ci-robot]

@qinqon: This pull request references CNV-40881 which is a valid jira issue.

Warning: The referenced jira issue has an invalid target version for the target branch this PR targets: expected the story to target either version "4.16." or "openshift-4.16.", but it targets "cnv-4.16" instead.

In response to this:

What this PR does / why we need it:
This change add a test to start a kubevirt hosted cluster with just secondary interface as primary network and start a companion dnsmasq pod attached to that network that acts as dhcp server, gateway and masquerade. The test only has one worker so this works as expected.

TODO:

• The default passthrough ingress service is pointing to the VM but default network cannot access that IP, either make it go over dnsmasq and redirect the https traffic to the VM there or connect the VM to the underlay (not sure if this would work though).

Which issue(s) this PR fixes (optional, use fixes #<issue_number>(, fixes #<issue_number>, ...) format, where issue_number might be a GitHub issue, or a Jira story:
Fixes #

Checklist

• Subject and description added to both, commit and PR.
• Relevant issues have been referenced.
• This change includes docs.
• This change includes unit tests.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the openshift-eng/jira-lifecycle-plugin repository.

[qinqon]

Before last changes we have a green lane now multiple nodepool tests are failing

 --- FAIL: TestNodePool/Main (0.02s)
        --- SKIP: TestNodePool/Main/TestRollingUpgrade (0.00s)
        --- SKIP: TestNodePool/Main/TestKMSRootVolumeEncryption (0.00s)
        --- SKIP: TestNodePool/Main/TestNodepoolMachineconfigGetsRolledout (0.00s)
        --- SKIP: TestNodePool/Main/TestNodePoolAutoRepair (0.00s)
        --- SKIP: TestNodePool/Main/TestNTOMachineConfigGetsRolledOut (0.00s)
        --- SKIP: TestNodePool/Main/TestNTOMachineConfigAppliedInPlace (0.00s)
        --- FAIL: TestNodePool/Main/KubeVirtNodeAdvancedMultinetTest (2814.02s)
        --- FAIL: TestNodePool/Main/TestNodePoolReplaceUpgrade (2815.09s)
        --- FAIL: TestNodePool/Main/KubeKubeVirtJsonPatchTest (2815.08s)
        --- FAIL: TestNodePool/Main/KubeVirtCacheTest (2820.09s)
        --- FAIL: TestNodePool/Main/TestNTOPerformanceProfile (2820.09s)
        --- FAIL: TestNodePool/Main/TestNodePoolInPlaceUpgrade (2835.08s)
        --- PASS: TestNodePool/Main/KubeVirtNodeSelectorTest (1845.14s)
        --- PASS: TestNodePool/Main/KubeVirtQoSClassGuaranteedTest (730.10s)
        --- PASS: TestNodePool/Main/KubeVirtNodeMultinetTest (485.11s)

hosted cluster is in partial state

=== NAME  TestNodePool/Main/KubeVirtNodeAdvancedMultinetTest
    util.go:340: 
        failed waiting for hostedcluster image rollout: status.version.history[0].state is "Partial", but we want "Completed"
        Unexpected error:
            <wait.errInterrupted>: 
            timed out waiting for the condition
            {
                cause: <*errors.errorString | 0xc000a10a90>{
                    s: "timed out waiting for the condition",
                },
            }
        occurred

[qinqon]

Passing locally and isolated

=== NAME  TestNodePool
    hypershift_framework.go:132: skipping teardown, already called
--- PASS: TestNodePool (688.64s)
    --- PASS: TestNodePool/Main (65.00s)
        --- PASS: TestNodePool/Main/KubeVirtNodeAdvancedMultinetTest (617.71s)
PASS
+ popd
~/Documents/cnv/sandbox/aws

[qinqon]

Running all Kubevirt nodepool in parallel we get the console operator stuck

console                                    4.16.0-0.nightly-2024-05-05-102537   False       False         True       7m      RouteHealthAvailable: failed to GET route (https://console-openshift-console.apps.example-tc26l.apps.cluster-1715061648-ellorent-test.devcluster.openshift.com): Get "https://console-openshift-console.apps.example-tc26l.apps.cluster-1715061648-ellorent-test.devcluster.openshift.com": context deadline exceeded (Client.Timeout exceeded while awaiting headers)

[qinqon]

It takes 20 seconds for the route to respond

time curl -Lk https://console-openshift-console.apps.example-tc26l.apps.cluster-1715061648-ellorent-test.devcluster.openshift.com
...
real	0m20.591s
user	0m0.020s
sys	0m0.011s

Even removing the endpointslices from from advanced multus we get that latency

[qinqon]

/retest

[qinqon]

Now is possible to run different nodepools at different hosted cluster and those are run in parallel too.

--- PASS: TestNodePool (0.00s)
    --- PASS: TestNodePool/HostedCluster1 (1123.09s)
        --- PASS: TestNodePool/HostedCluster1/Main (45.95s)
            --- PASS: TestNodePool/HostedCluster1/Main/KubeVirtNodeAdvancedMultinetTest (1066.60s)
    --- PASS: TestNodePool/HostedCluster0 (1163.22s)
        --- PASS: TestNodePool/HostedCluster0/Main (48.97s)
            --- PASS: TestNodePool/HostedCluster0/Main/KubeVirtNodeSelectorTest (1097.61s)
            --- PASS: TestNodePool/HostedCluster0/Main/KubeKubeVirtJsonPatchTest (1101.79s)
            --- PASS: TestNodePool/HostedCluster0/Main/KubeVirtNodeMultinetTest (1101.97s)
            --- PASS: TestNodePool/HostedCluster0/Main/KubeVirtQoSClassGuaranteedTest (1101.98s)
            --- PASS: TestNodePool/HostedCluster0/Main/KubeVirtCacheTest (1106.72s)

[openshift-ci-robot]

@qinqon: This pull request references CNV-40881 which is a valid jira issue.

Warning: The referenced jira issue has an invalid target version for the target branch this PR targets: expected the story to target either version "4.16." or "openshift-4.16.", but it targets "cnv-4.16" instead.

In response to this:

What this PR does / why we need it:
This change add a test to start a kubevirt hosted cluster with just secondary interface as primary network and start a companion dnsmasq pod attached to that network that acts as dhcp server, gateway and masquerade. The test only has one worker so this works as expected.

TODO:

• Replace quay.io/coreos/dnsmaq
• Skip HostedCluster1 is this is not kubevirt (aws lane should not run it)
• Move dnsmasq and nad back to the hosted cluster namespace and make it privileged.

Which issue(s) this PR fixes (optional, use fixes #<issue_number>(, fixes #<issue_number>, ...) format, where issue_number might be a GitHub issue, or a Jira story:
Fixes #

Checklist

• Subject and description added to both, commit and PR.
• Relevant issues have been referenced.
• This change includes docs.
• This change includes unit tests.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the openshift-eng/jira-lifecycle-plugin repository.

[openshift-ci-robot]

@qinqon: This pull request references CNV-40881 which is a valid jira issue.

Warning: The referenced jira issue has an invalid target version for the target branch this PR targets: expected the story to target either version "4.16." or "openshift-4.16.", but it targets "cnv-4.16" instead.

In response to this:

What this PR does / why we need it:
This change add a test to start a kubevirt hosted cluster with just secondary interface as primary network and start a companion dnsmasq pod attached to that network that acts as dhcp server, gateway and masquerade. The test only has one worker so this works as expected.

TODO:

• Replace quay.io/coreos/dnsmaq
• Don't start HotedCluster1 for test not running there.
• Use nodepool as owner of dnmasq and nad at default namespace.

Which issue(s) this PR fixes (optional, use fixes #<issue_number>(, fixes #<issue_number>, ...) format, where issue_number might be a GitHub issue, or a Jira story:
Fixes #

Checklist

• Subject and description added to both, commit and PR.
• Relevant issues have been referenced.
• This change includes docs.
• This change includes unit tests.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the openshift-eng/jira-lifecycle-plugin repository.

[openshift-ci-robot]

@qinqon: This pull request references CNV-40881 which is a valid jira issue.

Warning: The referenced jira issue has an invalid target version for the target branch this PR targets: expected the story to target either version "4.16." or "openshift-4.16.", but it targets "cnv-4.16" instead.

In response to this:

What this PR does / why we need it:
This change add a test to start a kubevirt hosted cluster with just secondary interface as primary network and start a companion dnsmasq pod attached to that network that acts as dhcp server, gateway and masquerade. The test only has one worker so this works as expected.

Which issue(s) this PR fixes (optional, use fixes #<issue_number>(, fixes #<issue_number>, ...) format, where issue_number might be a GitHub issue, or a Jira story:
Fixes #

Checklist

• Subject and description added to both, commit and PR.
• Relevant issues have been referenced.
• This change includes docs.
• This change includes unit tests.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the openshift-eng/jira-lifecycle-plugin repository.

[stevekuznetsov]

@qinqon re #3902 (comment) ... it's two fields (Compute, RootVolume) to filter, which you can do with a one-liner in cmpopts.IgnoreFields(hyperv1.KubevirtNodePoolPlatform{}, "Compute", "RootVolume"). Did you check what the output looks like when the test fails for real? That is the real issue - when the test fails in a way we expect, the output of Ginkgo is very, very hard to read. The structured diff out of cmp is simple.

I won't force you to but I really think you would benefit from comparing the output of the test when it fails the way you expect between the two approaches.

[qinqon]

cmpopts.IgnoreFields(hyperv1.KubevirtNodePoolPlatform{}, "Compute", "RootVolume")

This means we have to maintain this in case new fields are added, I think is not worth it.

[qinqon]

/test e2e-kubevirt-aws-ovn

Not related

--- FAIL: TestCreateCluster/Teardown (33.16s)
        --- FAIL: TestCreateCluster/Teardown/ValidateMetricsAreExposed (0.04s)

[openshift-ci]

@qinqon: The following tests failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name	Commit	Details	Required	Rerun command
ci/prow/e2e-azure	57f6fcc	link	false	/test e2e-azure
ci/prow/e2e-kubevirt-aws-ovn	57f6fcc	link	true	/test e2e-kubevirt-aws-ovn

Full PR test history. Your PR dashboard.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. I understand the commands that are listed here.