Supernets: AWS Terraform deployment problems

Developers Polygon CDK
1

Hi,

I’m attempting to follow the guide for Deploying Polygon Supernets on AWS with Terraform and I’m running into multiple issues.

I was able to deploy the infra described by Terraform, with the additional tweak of setting the jumpbox_ssh_access var to 0.0.0.0/0 (without this the ingres aws_security_group_rule.open_ssh creation times out).

However, when it comes to provisioning via Ansible, I’ve reached a blocker at the ansible -inventory --graph --inventory inventory/aws_ec2.yml command step. Firstly, the command turns out to be ansible-inventory (w/o the space, a doc update needed perhaps?), but when I run it, it does not produce an Ansible inventory file as stated by docs - I’m not familiar to Ansible, thus unsure how to troubleshoot.

Attempting the following command of pinging the instances also fails, times out actually, with the following:

i-05f5989624cecec38 | UNREACHABLE! => {
    "changed": false,
    "msg": "Failed to connect to the host via ssh: ssh: Could not resolve hostname i-05f5989624cecec38: nodename nor servname provided, or not known",
    "unreachable": true
}
i-03a4c372c32298ecf | UNREACHABLE! => {
    "changed": false,
    "msg": "Failed to connect to the host via ssh: ssh: Could not resolve hostname i-03a4c372c32298ecf: nodename nor servname provided, or not known",
    "unreachable": true
}
i-07d11061ec567f27a | UNREACHABLE! => {
    "changed": false,
    "msg": "Failed to connect to the host via ssh: ssh: Could not resolve hostname i-07d11061ec567f27a: nodename nor servname provided, or not known",
    "unreachable": true
}
i-048c975d430782359 | UNREACHABLE! => {
    "changed": false,
    "msg": "Failed to connect to the host via ssh: ssh: Could not resolve hostname i-048c975d430782359: nodename nor servname provided, or not known",
    "unreachable": true
}

Not stated in the docs, I also changed the filters:tag:BaseDN in aws_ec2.yml to correspond to the BaseDN output from the Terraform deployment - not sure if that is right, some docs coverage on this would be helpful.

Running ansible-inventory --graph --inventory inventory/aws_ec2.yml does produce this output:

@all:
  |--@ungrouped:
  |--@aws_ec2:
  |  |--i-03a4c372c32298ecf
  |  |--i-048c975d430782359
  |  |--i-05f5989624cecec38
  |  |--i-07d11061ec567f27a
  |--@validator:
  |  |--i-03a4c372c32298ecf
  |  |--i-048c975d430782359
  |  |--i-05f5989624cecec38
  |  |--i-07d11061ec567f27a
  |--@dev_edge_rg_private:
  |  |--i-03a4c372c32298ecf
  |  |--i-048c975d430782359
  |  |--i-05f5989624cecec38
  |  |--i-07d11061ec567f27a
  |--@validator_004:
  |  |--i-03a4c372c32298ecf
  |--@validator_002:
  |  |--i-048c975d430782359
  |--@validator_003:
  |  |--i-05f5989624cecec38
  |--@validator_001:
  |  |--i-07d11061ec567f27a

so I feel it’s close, but that I’m missing something.

Deepest gratitude in advance to any help!

Yuri

3

If you’re running into issues with Supernets while deploying AWS resources using Terraform, you’re not alone. A few common problems tend to come up:

  1. CIDR block overlaps – Make sure the supernet you’re defining doesn’t overlap with existing VPCs or subnets. Terraform won’t allow overlapping IP ranges, and AWS will throw an error.

  2. Subnet sizing – When splitting a supernet into smaller subnets, double-check that the CIDR mask sizes match your planned resources. Too small or too large can break deployment.

  3. Terraform state conflicts – If you manually changed subnet configurations in AWS after Terraform deployment, the state file can go out of sync. Running terraform import or terraform refresh can help align the state.

  4. Dependencies – Ensure that resources depending on these subnets (like EC2, RDS, or route tables) are declared properly in Terraform so they’re created in the right order.

Tips:

  • Use the cidrsubnet() function in Terraform carefully to generate subnets from a supernet.

  • Always validate the Terraform plan with terraform plan before applying.

  • Consider using separate modules for VPCs and subnets to make your deployments cleaner and easier to debug.

In short, most supernet deployment problems in Terraform come down to CIDR conflicts, state mismatches, or resource dependency issues. Debug step by step, and Terraform will usually give clear hints on what’s wrong.