02/20/2023, 2:40 PM
Hey guys, Im deploying an EKS Cluster that after the creation it generates a kubeconfig file and uses it to create a k8s provider. after the cluster and the node group is finished and the provider is live, the cluster begins installing deployments using pulumi helm v3 release. for some reason, the deployments keep failing randomly( sometimes a couple are ok and other are not and vice versa) for this reason:
Helm release "cert-manager/cert-manager" was created, but failed to initialize completely. Use Helm CLI to investigate.: failed to become available within allocated timeout. Error: Helm Release cert-manager/cert-manager: the server could not find the requested resource
after retrying with pulumi up everything is fine and it deploys as shown in the picture. help anyone please? 🙂 *I tried using every flag possible - skipAwait, waitForJobs, timeout (respectively) and neither are working


02/20/2023, 7:52 PM
are you positive that you are installing the chart only after the worker nodes are ready? Because it's possible you are installing right after the control plane is ready but before the worker nodes are, so there is nowhere to place the pods


02/21/2023, 3:49 PM
likely need to add a depends on to the helm release for the node group
can only infer the order and dependencies on whether you pass outputs from one to the other and on the explict depends on