No matter how you like to participate in developer communities, Pulumi wants to meet you there. If you want to meet other Pulumi users to share use-cases and best practices, contribute code or documentation, see us at an event, or just tell a story about something cool you did with Pulumi, you are part of our community.

Pulumi Community

Hey, I'm using Pulumi to deploy an EKS cluster and install some Helm charts on it. 
Everything is working when I run Pulumi using the CLI on my laptop but since last week, Pulumi preview stopped working on Github actions with the following error: “kubernetes: <http://helm.sh/v3:Release|helm.sh/v3:Release> (helm-chart-name): error: could not get server version from Kubernetes: the server has asked for the client to provide credentials”. Any clues?

could you double check your K8s provider and make sure it's not using outdated credentials

Thanks, already checked that. The k8s provider kubeconfig is ok, and i can run a preview from my local using the cli but from github actions fails

does the github runner itself still have permissions to authenticate to the cluster? e.g. if you run a plain `kubectl get po` command from the runner does it succeed?

Yes, i can run that command and get the output

can you share your provider configuration and helm code?

```kubeConfig, err := cluster.GetKubeconfig(ctx, &amp;eks.ClusterGetKubeconfigArgs{
			ProfileName: pulumi.String(profileName),
		})
		if err != nil {
			return nil, err
		}

k8sProvider, err := kubernetes.NewProvider(ctx, clusterResName, &amp;kubernetes.ProviderArgs{
			Kubeconfig: kubeConfig,
		})
		if err != nil {
			return nil, err
		}

_, err := helm.NewRelease(ctx, name, &amp;helm.ReleaseArgs{
			Chart:     pulumi.String(chart.Name),
			Version:   pulumi.String(chart.Version),
			Namespace: pulumi.String(chart.Namespace),
			RepositoryOpts: helm.RepositoryOptsArgs{
				Repo: pulumi.String(chart.Repo),
			}
		}, pulumi.Provider(k8sProvider))
		if err != nil {
			return err
		}```

I think it is something related to how aws profiles are managed, but not sure what the problem is yet

if the kubeConfig specifies an AWS_PROFILE then your runner will need that same profile to be available. Or it will need to be excluded from the provider you created

yeah, I have a profile with the same name on github actions