Failure to run on AKS cluster about kubescape HOT 14 CLOSED

@EtienneDeneuve

from kubescape.

@muandane we have just released a new Chart to address this issue, can you try again?

from kubescape.

1.18.3 released 2 minutes ago

from kubescape.

thanks @matthyx i dont get that erreur anymore, but i'm getting this instead is it because of kubescape not being fully compatible with AKS ?

from kubescape.

does it have to do with 1 kubevuln pod being down due to ephemeral-storage limitation ?
Pod ephemeral local storage usage exceeds the total limit of containers 4Gi

from kubescape.

thanks @matthyx i dont get that erreur anymore, but i'm getting this instead is it because of kubescape not being fully compatible with AKS ?

It means we didn't detect you're running on AKS... you can try setting cloudProviderMetadata.cloudProviderEngine=aks

from kubescape.

does it have to do with 1 kubevuln pod being down due to ephemeral-storage limitation ? `Pod ephemeral local storage

kubevuln doesn't have to do with compliance scan, it's the component that does vulnerability scans... however it would be nice to have it running :)
on your cluster you don't have PV support?

from kubescape.

I'm installing kubescape using flux, this is the values structure i'm giving flux

ps:

• i have also added the secrets to the kuebscape deployment but in a form a secret instead of passing them as stings using envFrom.secretRef
• for the PV yes my aks cluster supports it i added as you can see the storage class needed
• for the cloudProviderMetadata.cloudProviderEngine=aks i checked the chart locally and it's not used anywhere i think that's the issue

from kubescape.

right, this parameter was deprecated looooong ago... sorry
maybe you could try setting some of the AKS specific values cloudProviderMetadata.aks* from https://github.com/kubescape/helm-charts/tree/main/charts/kubescape-operator ?

from kubescape.

yes i added all of them as a secret

from kubescape.

hmm, @dwertent any idea?

from kubescape.

I think this is because we dont get the cloud provider from the nodes.
I will change this behavior.

from kubescape.

I revisited the issue and found that Kubescape identifies the cloud provider.
Have you reviewed the documentation here? It mentions both User-Assigned Managed Identities and System-Assigned Managed Identities. Ensure that you have defined the correct one.

from kubescape.

@dwertent I think your recent fix in the latest release should solve that.

from kubescape.