step:
  name: "Deploy to prod environment"
  image: atlassian/default-image:2
  script:
    - source .env
    - envsubst < k8s/deploy.yaml > k8s/deploy-apply.yaml
    - cat k8s/deploy-apply.yaml
    - pipe: atlassian/aws-eks-kubectl-run:2.2.1
      variables:
        AWS_ACCESS_KEY_ID: $AWS_ACCESS_KEY_ID
        AWS_SECRET_ACCESS_KEY: $AWS_SECRET_ACCESS_KEY
        AWS_DEFAULT_REGION: $AWS_DEFAULT_REGION
        CLUSTER_NAME: $CLUSTER_NAME
        KUBECTL_COMMAND: "apply"
        RESOURCE_PATH: "k8s/deploy-apply.yaml"
        DEBUG: "true"

字符串
我已经检查了k8s/deploy-apply.yaml，当手动应用时，它运行成功，在替换下面的变量后，命令运行成功：

kubectl apply -f k8s/deploy-apply.yaml

型
但是在bitbucket-pipeline中运行时，我得到以下错误：

INFO: Successfully updated the kube config.
error: You must be logged in to the server (the server has asked for the client to provide credentials)
✖ spec file validation failed.

型
我已经检查了AWS_ACCESS_KEY_ID，AWS_SECRET_ACCESS_KEY，它们是有效的，有权限。