AWSResourceReference unimplemented logic

[MarcusNoble]

Collating all instances of AWSResourceReference being used in the codebase without implementing all lookup logic (e.g. only using the ID property)

Type	Property	References	ID	ARN	Filters
AWSMachineSpec	AMI	cluster-api-provider-aws/pkg/cloud/services/ec2/instances.go Lines 132 to 134 in 2e24bd1 if scope.AWSMachine.Spec.AMI.ID != nil { // nolint:nestif input.ImageID = *scope.AWSMachine.Spec.AMI.ID } else {	✅	❌	❌
AWSMachineSpec	AdditionalSecurityGroups	cluster-api-provider-aws/controllers/awsmachine_security_groups.go Lines 62 to 64 in 2e24bd1 for _, id := range additional { newAnnotation[*id.ID] = struct{}{} }	✅	❌	❌
AWSMachineSpec	Subnet	cluster-api-provider-aws/pkg/cloud/services/ec2/instances.go Lines 261 to 343 in 2e24bd1 switch { case scope.AWSMachine.Spec.Subnet != nil && scope.AWSMachine.Spec.Subnet.ID != nil: if failureDomain != nil { subnet := s.scope.Subnets().FindByID(*scope.AWSMachine.Spec.Subnet.ID) if subnet == nil { record.Warnf(scope.AWSMachine, "FailedCreate", "Failed to create instance: subnet with id %q not found", aws.StringValue(scope.AWSMachine.Spec.Subnet.ID)) return "", awserrors.NewFailedDependency( fmt.Sprintf("failed to run machine %q, subnet with id %q not found", scope.Name(), aws.StringValue(scope.AWSMachine.Spec.Subnet.ID), ), ) } if subnet.AvailabilityZone != *failureDomain { record.Warnf(scope.AWSMachine, "FailedCreate", "Failed to create instance: subnet's availability zone %q does not match with the failure domain %q", subnet.AvailabilityZone, *failureDomain) return "", awserrors.NewFailedDependency( fmt.Sprintf("failed to run machine %q, subnet's availability zone %q does not match with the failure domain %q", scope.Name(), subnet.AvailabilityZone, *failureDomain, ), ) } } return *scope.AWSMachine.Spec.Subnet.ID, nil case scope.AWSMachine.Spec.Subnet != nil && scope.AWSMachine.Spec.Subnet.Filters != nil: criteria := []*ec2.Filter{ filter.EC2.SubnetStates(ec2.SubnetStatePending, ec2.SubnetStateAvailable), filter.EC2.VPC(s.scope.VPC().ID), } if failureDomain != nil { criteria = append(criteria, filter.EC2.AvailabilityZone(*failureDomain)) } for _, f := range scope.AWSMachine.Spec.Subnet.Filters { criteria = append(criteria, &ec2.Filter{Name: aws.String(f.Name), Values: aws.StringSlice(f.Values)}) } subnets, err := s.getFilteredSubnets(criteria...) if err != nil { return "", errors.Wrapf(err, "failed to filter subnets for criteria %q", criteria) } if len(subnets) == 0 { record.Warnf(scope.AWSMachine, "FailedCreate", "Failed to create instance: no subnets available matching filters %q", scope.AWSMachine.Spec.Subnet.Filters) return "", awserrors.NewFailedDependency( fmt.Sprintf("failed to run machine %q, no subnets available matching filters %q", scope.Name(), scope.AWSMachine.Spec.Subnet.Filters, ), ) } return *subnets[0].SubnetId, nil case failureDomain != nil: subnets := s.scope.Subnets().FilterPrivate().FilterByZone(*failureDomain) if len(subnets) == 0 { record.Warnf(scope.AWSMachine, "FailedCreate", "Failed to create instance: no subnets available in availability zone %q", *failureDomain) return "", awserrors.NewFailedDependency( fmt.Sprintf("failed to run machine %q, no subnets available in availability zone %q", scope.Name(), *failureDomain, ), ) } return subnets[0].ID, nil // TODO(vincepri): Define a tag that would allow to pick a preferred subnet in an AZ when working // with control plane machines. default: sns := s.scope.Subnets().FilterPrivate() if len(sns) == 0 { record.Eventf(s.scope.InfraCluster(), "FailedCreateInstance", "Failed to run machine %q, no subnets available", scope.Name()) return "", awserrors.NewFailedDependency(fmt.Sprintf("failed to run machine %q, no subnets available", scope.Name())) } return sns[0].ID, nil }	✅	❌	✅
AWSMachinePoolSpec	Subnets	cluster-api-provider-aws/pkg/cloud/services/autoscaling/autoscalinggroup.go Lines 144 to 146 in 2e24bd1 for i, v := range scope.AWSMachinePool.Spec.Subnets { subnetIDs[i] = aws.StringValue(v.ID) }	✅	❌	❌

[MarcusNoble]

It looks as though there aren't as many instances as first thought.

If ignoring test files and generated files AWSResourceReference is only made use of in 4 places.

All 4 make use of the ID property, one makes use of the Filters property and there are no instances where ARN is used.

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-contributor-experience at kubernetes/community.
/lifecycle stale

[MarcusNoble]

/remove-lifecycle stale

[sedefsavas]

/assign @randomvariable

[randomvariable]

/lifecycle active

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle stale
• Mark this issue or PR as rotten with /lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle rotten

[k8s-triage-robot]

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Reopen this issue or PR with /reopen
• Mark this issue or PR as fresh with /remove-lifecycle rotten
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close

[k8s-ci-robot]

@k8s-triage-robot: Closing this issue.

In response to this:

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Reopen this issue or PR with /reopen
• Mark this issue or PR as fresh with /remove-lifecycle rotten
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close

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.

[sedefsavas]

lifecycle/stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle rotten

[Ankitasw]

/assign

[Ankitasw]

@sedefsavas @richardcase Just to clarify, here we need to implement lookup logic for fields which are of type AWSResourceReference and making use of any of the properties(ID, Filters) from AWSResourceReference in the respective implementation.

[Ankitasw]

/lifecycle active

[shivi28]

Pasting slack discussion here for others reference:
Scope of this issue has multiple parts, and one of them is

if subnet id is there, do a lookup using that
if filter is there, do a lookup using that
If both are given then throw a validation error 

But currently in our code we are only doing lookup by IDs.

So a PR is raised which will implement above logic for AWSResourceReference in AWSMachinePool.Spec.Subnets during ASG creation/updation.

/assign @shivi28