[Bug]: ElastiCache replication group enabled Global Datastore is unnecessarily waiting to be dissociated when it's destroyed

[yuji-paypay]

Terraform Core Version

5.62.0

AWS Provider Version

1.7.5

Affected Resource(s)

aws_elasticache_global_replication_group

elasticache_replication_group

Expected Behavior

The destruction should proceed without waiting for the dissociation of replica from Global DataStore after it's already dissociated

Actual Behavior

During destruction, it waits for dissociation of replica from Global Datastore and it just ends up timeout.

Relevant Error/Panic Output Snippet

No response

Terraform Configuration Files

• resource "aws_elasticache_global_replication_group" "elasticache_global_replication_group" {

  • global_replication_group_description = "test-redis7 redis cluster replication group"
  • global_replication_group_id_suffix = "test-redis7" }

• resource "aws_elasticache_replication_group" "primary-redis-cluster-encrypted" {

  • apply_immediately = true
  • at_rest_encryption_enabled = true
  • auth_token_update_strategy = "ROTATE"
  • auto_minor_version_upgrade = "false"
  • automatic_failover_enabled = true
  • description = "test-redis7 redis cluster"
  • engine = "redis"
  • engine_version = "7.0"
  • maintenance_window = "sun:18:00-sun:19:00"
  • multi_az_enabled = true
  • node_type = "cache.m5.large"
  • num_node_groups = 1
  • port = 6379
  • replicas_per_node_group = 1
  • replication_group_id = "test-redis7"
  • snapshot_retention_limit = 2
  • snapshot_window = "17:00-18:00"
  • subnet_group_name = "subnet-group"
  • transit_encryption_enabled = true }

• resource "aws_elasticache_replication_group" "replica-redis-cluster-encrypted" {

  • apply_immediately = true
  • auth_token_update_strategy = "ROTATE"
  • automatic_failover_enabled = true
  • description = "test-redis7 redis cluster"
  • engine = "redis"
  • maintenance_window = "sun:18:00-sun:19:00"
  • multi_az_enabled = true
  • port = 6379
  • replicas_per_node_group = 1
  • replication_group_id = "test-redis7"
  • snapshot_retention_limit = 2
  • snapshot_window = "17:00-18:00"
  • subnet_group_name = "subnet-group" }

Steps to Reproduce

• Create the replication groups in each region with the same ID
• Destroy global datastore

Debug Output

No response

Panic Output

No response

Important Factoids

No response

References

Looks like there is no consideration about region. So it may be waiting for the primary when it is dissociating replica from global datastore https://github.com/hashicorp/terraform-provider-aws/blob/main/internal/service/elasticache/global_replication_group.go#L787

Actually I resolved this issue to have different ID between primary and replica

Would you like to implement a fix?

None

[github-actions[bot]]

Community Note

Voting for Prioritization

• Please vote on this issue by adding a 👍 reaction to the original post to help the community and maintainers prioritize this request.
• Please see our prioritization guide for information on how we prioritize.
• Please do not leave "+1" or other comments that do not add relevant new information or questions, they generate extra noise for issue followers and do not help prioritize the request.

Volunteering to Work on This Issue

• If you are interested in working on this issue, please leave a comment.
• If this would be your first contribution, please review the contribution guide.