Setting up a standby cluster with AWS

[DadiAnas]

In case you are using AWS, please make sure that the AWS user you provided its credentials with STANDBY_AWS_ACCESS_KEY_ID, STANDBY_AWS_SECRET_ACCESS_KEY and STANDBY_AWS_REGION has permissions to deal with s3, and that you have created a role with following policies:

Policy:

{
    "Statement": [
        {
            "Action": [
                "ec2:Describe*",
                "ec2:Describe*",
                "ec2:ModifyVolumeAttribute"
            ],
            "Effect": "Allow",
            "Resource": "*"
        }
    ],
    "Version": "2012-10-17"
}

Role:

{
    Version = "2012-10-17"
    Statement = [
      {
        Action = "sts:AssumeRole"
        Effect = "Allow"
        Sid    = ""
        Principal = {
          Service = "ec2.amazonaws.com"
        }
      },
    ]
}

Then, add the role name to kube_iam_role param or the annotation iam.amazonaws.com/role: "postgres-operator-role" to the operator deployment.

[FxKu]

Always open to improve docs. This would be better suited for the admin docs, I believe. Can you check this paragraph if something can be added there? I would prefer yaml over json examples. We do not list the policy yet.

For the user docs it might make sense to add just one sentence that cloning / standby from s3 only works if continuous archiving is set up correctly.