Terraform code for Leverage Reference Architecture for AWS, designed under optimal configs for the most popular modern web and mobile applications needs.
When running a layer with a Kubernetes provider, this error arises randomly:
╷
│ Error: Invalid credentials
│
│ with kubernetes_manifest.mixed_routing_ingress,
│ on mixed-routing-ingress.tf line 1, in resource "kubernetes_manifest" "mixed_routing_ingress":
│ 1: resource "kubernetes_manifest" "mixed_routing_ingress" {
│
│ The credentials configured in the provider block are not accepted by the
│ API server. Error: Unauthorized
│
│ Set TF_LOG=debug and look for '[InvalidClientConfiguration]' in the log to
│ see actual configuration.
╵
Expected Behavior
The Kubernetes resources are applied with no issue.
Steps to Reproduce
Steps to reproduce the behavior:
Go to a layer with Kubernetes provider
Run leverage tf plan or leverage tf apply
Some times this error happens (specially in a CI pipeline)
Screenshots
N/A
Environment (please complete the following information):
Describe the Bug
When running a layer with a Kubernetes provider, this error arises randomly:
Expected Behavior
The Kubernetes resources are applied with no issue.
Steps to Reproduce
Steps to reproduce the behavior:
leverage tf plan
orleverage tf apply
Screenshots
N/A
Environment (please complete the following information):
The error was seen in Gitlab CI pipelines.
Additional Context
As per these posts:
And this Kubernetes API specification: https://kubernetes.io/docs/reference/config-api/client-authentication.v1/
...a possible cause is the token not being updated here:
If this is the case, it has to be confirmed, the possible solution is to use some way to force the token update, e.g. replace the token line with:
So, the full provider definition is:
Today (2024-01-26) this is being tested in a client using binbash Leverage.