Resource 'aws_elasticache_cluster.redis' does not have attribute 'cache_nodes.*.address' for variable 'aws_elasticache_cluster.redis.cache_nodes.*.address'

[iroller]

Terraform Version

0.7.0

Affected Resource(s)

• aws_elasticache_cluster
• module outputs

Terraform Configuration Files

## Module
resource "aws_elasticache_cluster" "redis" {
  engine               = "redis"
...
}

// Addresses
output "addresses" {
  value = "${join(",", aws_elasticache_cluster.redis.cache_nodes.*.address)}"
}


## Using module
module "redis" {
...
}

output "redis_endpoint_addresses" {
  value = "${module.redis.addresses}"
}

Debug Output

[WARN] Output interpolation "addresses" failed: Resource 'aws_elasticache_cluster.redis' does not have attribute 'cache_nodes.*.address' for variable 'aws_elasticache_cluster.redis.cache_nodes.*.address'

Expected Behavior

I expected a string outout, e.g. "10.10.0.1,10.10.0.2"

Actual Behavior

No output at all.

References

• Resource 'aws_elasticache_cluster.redis' does not have attribute 'cache_nodes' for variable 'aws_elasticache_cluster.redis.*.cache_nodes' #4703 - similar, but not sure if it's the same issue

[iroller]

The same happens when using memcached engine:

"${formatlist("%s:%s", aws_elasticache_cluster.memcached.cache_nodes.*.address, aws_elasticache_cluster.memcached.cache_nodes.*.port)}"

^^gives no output

[sumerman]

I had the same issue with my plugin and this is my workaround.

Essentially instead of using endpoint.address as a source for the DNS record, I've added a surrogate attribute endpoint_address which works just fine. This leads me to believe there is an issue with nested computed attributes in core.

P.S.: to make a repro case out my test it's enough to replace endpoint_address with endpoint.address here

[sumerman]

I believe this issue is related to #7241 and/or #3679

[steblynskyi]

records = ["${aws_elasticache_cluster.session-storage.cache_nodes.0.address}"]

depends_on = ["aws_elasticache_cluster.session-storage"]

[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 have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.