BREAKING CHANGES: To add add SPOT support for MNG, the `instance_type` is now a list and renamed as `instance_types`. This will probably rebuild existing Managed Node Groups.
NOTES: The EKS cluster can be provisioned with both private and public subnets. But Fargate only accepts private ones. This new variable allows to override the subnets to explicitly pass the private subnets to Fargate and work around that issue.
NOTES: Tags that are passed into `var.worker_groups_launch_template` or `var.worker_groups` now override tags passed in via `var.tags` for Autoscaling Groups only. This allow ASG Tags to be overwritten, so that `propagate_at_launch` can be tweaked for a particular key.
NOTES: The output `cloudwatch_log_group_name` was incorrectly returning the log group name as a list of strings. As a workaround, people were using `module.eks_cluster.cloudwatch_log_group_name[0]` but that was totally inconsistent with output name. Those users can now use `module.eks_cluster.cloudwatch_log_group_name` directly.
NOTES: Keep in mind that changing the order of workers group is a destructive operation. All workers group are destroyed and recreated. If you want to do this safely, you should move then in state with `terraform state mv` until we manage workers groups as maps.
NOTE: The usage of customer managed policy, not an inline policy, for the `cluster_elb_sl_role_creation policy` is common for "enterprise" AWS users to disallow inline policies with an SCP rule for auditing-related reasons, and this accomplishes the same thing.
Additional support for Terraform v0.13 and aws v3!
- The update to the vpc module in examples was, strictly speaking, unnecessary but it adds the terraform block with supported versions.
- Update for iam module in the example was very necessary to support new versions
- Workaround for "Provider produced inconsistent final plan" when creating ASGs at the same time as the cluster. See https://github.com/terraform-providers/terraform-provider-aws/issues/14085 for full details.
- Blacklist 0.13.0 as it was too strict when migrating from aws v2 to v3 about dropped attributes.
BREAKING CHANGES: Default for `cluster_endpoint_private_access_cidrs` is now `null` instead of `["0.0.0.0/0"]`. It makes the variable required when `cluster_create_endpoint_private_access_sg_rule` is set to `true`. This will force everyone who want to have a private access to set explicitly their allowed subnets for the sake of the principle of least access by default.
* Add example for lauch config with mixed lifecycles
* Set what on-demand instance is
* Tweak wording
Co-authored-by: Thomas O'Neill <toneill@new-innov.com>
Co-authored-by: Daniel Piddock <daniel.piddock@teamcmp.com>
