expected type *v1beta1.Deployment, but watch event object had type *v1.Deployment

[gjcarneiro]

Is this a BUG REPORT or FEATURE REQUEST?:

Uncomment only one, leave it on its own line:

/kind bug

/kind feature

What happened:

Deployed quay.io/coreos/kube-state-metrics:v1.6.0-rc.0

I keep getting these messages in logs, over and over:

E0410 15:22:20.836131       1 reflector.go:329] k8s.io/kube-state-metrics/internal/collector/builder.go:482: expected type *v1beta1.Deployment, but watch event object had type *v1.Deployment
E0410 15:22:20.842987       1 reflector.go:329] k8s.io/kube-state-metrics/internal/collector/builder.go:482: expected type *v1beta1.Deployment, but watch event object had type *v1.Deployment
E0410 15:22:20.876726       1 reflector.go:329] k8s.io/kube-state-metrics/internal/collector/builder.go:482: expected type *v1beta1.Deployment, but watch event object had type *v1.Deployment

What you expected to happen:

It shouldn't log errors.

How to reproduce it (as minimally and precisely as possible):

This my deployment yaml file:

kube-state-metrics.yaml.txt

Anything else we need to know?:

Environment:

• Kubernetes version (use kubectl version): v1.13.3
• Kube-state-metrics image version v1.6.0-rc.0

[tariq1890]

Thanks for reporting this issue. We used extensions apiGroup for the sake of supporting older versions of k8s.

If we are to move the apiGroup to core v1 , we would need to look into updating the compatibility of matrix of kube-state-metrics v1.6.0.

Thoughts @brancz?

[brancz]

Is that so? Workloads went v1 in 1.9 if I recall correctly, so our compatibility matrix is correct as is (it says we don't even make any support claims about 1.9 at this point). Is the extensions API still available in the latest versions?

[tariq1890]

So I checked release-1.14. The extensions API is still there but it does say that extensions/v1beta1 is deprecated in favour of apps/v1beta2/ https://github.com/kubernetes/api/blob/d687e77c8ae999eab9ebad49439d16e61add0df0/extensions/v1beta1/types.go#L82

It doesn't seem to explain this error though. Will have to dig deeper.

[tariq1890]

Please disregard my earlier comments.

This should be fixed by #720

[tariq1890]

I think this can be closed now.