Allow creating MultiAZ memcached cluster in a single terraform apply where number nodes > number of AZ's

[hashibot]

This issue was originally opened by @msroest as hashicorp/terraform#7850. It was migrated here as part of the provider split. The original body of the issue is below.

---

Currently a aws_elasticache_cluster using MultiAZ setup with num_cache_nodes > then count of availability_zones.

AWS validation will fail with the following error:
aws_elasticache_cluster.swimlane-memcache: Error creating Elasticache: InvalidParameterCombination: Must specify the same number of preferred availability zones as requested number of nodes.
status code: 400, request id: 63124b02-54f6-11e6-95b5-3960d6e4870c

It can be done in 2 stages by setting num_cache_nodes to the same as the number of AZ's then after creation is complete change the num_cache_nodes to the final desired number.

It would be nice if terraform would do this itself.

[michael-henderson]

This would definitely be nice to have.

[bflad]

Pull request submitted: #4741

[bflad]

The aws_elasticache_cluster resource in version 1.23.0 of the AWS provider will have the current availability_zones argument (TypeSet) deprecated in favor of a new preferred_availability_zones argument (TypeList). This will allow you to correctly specify duplicate elements in the list as necessary.

[bflad]

This has been released in version 1.23.0 of the AWS provider. Please see the Terraform documentation on provider versioning or reach out if you need any assistance upgrading.

[ghost]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.

If you feel this issue should be reopened, we encourage creating a new issue linking back to this one for added context. Thanks!