Defining more cache nodes than you have availability zones is not supported

[gusse]

Found a bug? Maybe our Slack Community can help.

Describe the Bug

You cannot define more cache nodes than you have defined availability zones.

Expected Behavior

You should be able to have more nodes than AZs (max 20 for memcached), terraform documentation does mention that the preferred_availability_zones parameter should have a list with same amount of AZ as you have num_cache_nodes defined (https://registry.terraform.io/providers/hashicorp/aws/latest/docs/resources/elasticache_cluster#preferred_availability_zones) so I do understand the attempt to use slice() function in https://github.com/cloudposse/terraform-aws-elasticache-memcached/blob/master/main.tf#L101, but I doesn't work for this use case.

Steps to Reproduce

Steps to reproduce the behavior:

1. Define more nodes than AZs, e.g. 2 AZs and 3 nodes.

module "memcached" {
  source             = "git::https://github.com/cloudposse/terraform-aws-elasticache-memcached.git?ref=tags/0.9.2"
  name               = "test"
  availability_zones = ["us-east-1a", "us-east-1b"]
  vpc_id             = "vpc-123456"
  cluster_size       = 3
  instance_type      = "cache.r5.large"
  engine_version     = "1.5.16"
}

2. Run terraform plan
3. See error

Error: Invalid function argument

  on .terraform/modules/memcached/main.tf line 101, in resource "aws_elasticache_cluster" "default":
 101:   preferred_availability_zones = slice(var.availability_zones, 0, var.cluster_size)
    |----------------
    | var.cluster_size is 3

Invalid value for "end_index" parameter: end index must not be greater than
the length of the list.

Additional Context

I'll try to figure out a way to construct a list with AZs that matches the number of cache nodes and make a PR myself.

[gusse]

Oh yea, I guess you could define the same AZ multiple times in the var.availability_zones list, like follows:

...
  availability_zones = ["us-east-1a", "us-east-1b", "us-east-1c" ,"us-east-1d", "us-east-1e", "us-east-1f", "us-east-1a"]
  cluster_size       = 7
...

But that is a workaround..