Bug 1964540: Extend trunk configuration to port level in machineset

[adduarte]

Extend trunk configuration to port level in machineset
With the introduction of new technologies, like SR-IOV,
it is required that the user have the ability to
configured trunked and not-trunked ports for the same machines
using spec.ports.trunk.

• Extended trunk support needed check to cover spec.port and spec.network
• Makes ports inherit trunk setting from spec.Trunk if not specified.
• Refactored trunking to its own function. Trunk name matches port name
• Added trunking to spec.ports

OSASINFRA-2504

[adduarte]

As it currently stands, this patch brings us up to par with kubernetes-sigs/cluster-api-provider-openstack functionality:
you can define interfaces under "network" and "port" and you can trunk All of them or None of them.

This make a deployment using kuryr but with SR-IOV ports not trunked impossible.

[adduarte]

/retest

[adduarte]

This functionality is being addressed upstream in pr kubernetes-sigs#934.
Is best if we rework this pr to mirror upstream code and logic.

[dulek]

Looks good to me, but I'll leave the lgtm label for folks that are more familiar with this code.

[adduarte]

/uncc Fedosin
/cc mandre

[adduarte]

/approve (just to be explicit)

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: adduarte

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 [adduarte]

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

[mandre]

I think it should work - CI at least shows there is no regression. Were you able to test this patch?

[adduarte]

I should have enough testing done locally to verify its functionality soon.

[adduarte]

@mandre just spend the day testing and it seems to work as expected. Made some trivial changes to name trunks consistently.
But I tested
network.trunk = true & port.trunk=(nil|false,true) and network.trunk=false & port.trunk=(nil,false,true) and all seems to work as expected.

[openshift-ci]

@adduarte: This pull request references Bugzilla bug 1964540, which is invalid:

• expected the bug to target the "4.9.0" release, but it targets "---" instead

Comment /bugzilla refresh to re-evaluate validity if changes to the Bugzilla bug are made, or edit the title of this pull request to link to a different bug.

In response to this:

Bug 1964540: Extend trunk configuration to port level in machineset

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.

[mandre]

/bugzilla refresh

[openshift-ci]

@mandre: This pull request references Bugzilla bug 1964540, which is valid. The bug has been moved to the POST state. The bug has been updated to refer to the pull request using the external bug tracker.

3 validation(s) were run on this bug

• bug is open, matching expected state (open)
• bug target release (4.9.0) matches configured target release for branch (4.9.0)
• bug is in the state NEW, which is one of the valid states (NEW, ASSIGNED, ON_DEV, POST, POST)

Requesting review from QA contact:
/cc @eurijon

In response to this:

/bugzilla 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 kubernetes/test-infra repository.

[openshift-ci]

@openshift-ci[bot]: GitHub didn't allow me to request PR reviews from the following users: eurijon.

Note that only openshift members and repo collaborators can review this PR, and authors cannot review their own PRs.

In response to this:

@mandre: This pull request references Bugzilla bug 1964540, which is valid. The bug has been moved to the POST state. The bug has been updated to refer to the pull request using the external bug tracker.

3 validation(s) were run on this bug

• bug is open, matching expected state (open)
• bug target release (4.9.0) matches configured target release for branch (4.9.0)
• bug is in the state NEW, which is one of the valid states (NEW, ASSIGNED, ON_DEV, POST, POST)

Requesting review from QA contact:
/cc @eurijon

In response to this:

/bugzilla 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 kubernetes/test-infra repository.

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.

[adduarte]

Addressed comment about trunkName. See inline reply

[adduarte]

@dulek could you take a ,ook at the code that sets the name of the trunk.
Perhaps we shoudl not change the name of the trunk at all...

[iamemilio]

we should probably wait for the upstream patch to settle and merge before approving this, it seems like they are still addressing some comments.

[mandre]

we should probably wait for the upstream patch to settle and merge before approving this, it seems like they are still addressing some comments.

Keep in mind we're talking about downstream CAPO and we want to retire this code soon anyway. We don't have to wait for the upstream code to merge (we'll inherit it anyway once we move to MAPO), however we need to make absolutely sure the API is stable and agreed on upstream. This is what we must implement in this patch.

[adduarte]

the current version of this patch mimics upstream:
trunks are created with trunkname=port.name,

[iamemilio]

Keep in mind we're talking about downstream CAPO and we want to retire this code soon anyway. We don't have to wait for the upstream code to merge (we'll inherit it anyway once we move to MAPO), however we need to make absolutely sure the API is stable and agreed on upstream. This is what we must implement in this patch.

true, as long as the upstream api is stable, I suppose its ok

[iamemilio]

little nit pick, but otherwise lgtm

[iamemilio]

/lgtm

[openshift-ci]

@adduarte: All pull requests linked via external trackers have merged:

• openshift/cluster-api-provider-openstack#185

Bugzilla bug 1964540 has been moved to the MODIFIED state.

In response to this:

Bug 1964540: Extend trunk configuration to port level in machineset

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.