Connecting to docker via ssh results in host key verification error

[mavogel]

This issue was originally opened by @yellowmegaman as https://github.com/hashicorp/terraform-provider-docker/issues/170. It was migrated here as a result of the community provider takeover from @kreuzwerker. The original body of the issue is below.

---

Terraform Version

Terraform v0.12.5 terraform-providers/docker 2.1.0

Affected Resource(s)

• docker provider itself

Terraform Configuration Files

provider "docker" {
  host = "ssh://cloud@${google_compute_instance.testinstance.0.network_interface.0.access_config.0.nat_ip}:22"
}

resource "google_compute_instance" "testinstance" {
  allow_stopping_for_update = true
  count        = "1"
  name         = "testinstance"
  machine_type = "n1-standard-1"
  zone         = "europe-west3-a"
  boot_disk {
    initialize_params {
      size  = "30"
    }
  }
  service_account {
    scopes = ["compute-rw", "storage-ro"]
  }
  network_interface {
    network = "default"
    access_config {
      nat_ip = google_compute_address.testinstance-static-ip-address.address
    }
  }
}

• valid key is present on the machine
• can ssh to target host with docker daemon listening on 2375

Debug Output

...
Error: Error pinging Docker server: error during connect: Get http://docker/_ping: command [ssh -l cloud -p 22 XX.XX.XX.XXX -- docker system dial-stdio] has exited with exit status 255, please make sure the URL is valid, and Docker 18.09 or later is installed on the remote host: stderr=Host key verification failed.
on 69.tf line 3, in provider "docker":
3: provider "docker" {
2019-07-30T19:54:55.369Z [DEBUG] plugin: plugin process exited: path=/drone/src/.terraform/plugins/linux_amd64/terraform-provider-docker_v2.1.0_x4 pid=206
2019-07-30T19:54:55.369Z [DEBUG] plugin: plugin exited

Expected Behavior

Successful connection to docker daemon

Actual Behavior

Connection error

Steps to Reproduce

Please list the steps required to reproduce the issue, for example:

1. Machine already created
2. Add docker configuration
3. terraform apply

Thanks a bunch in advance!

[aleon1220]

I am very interested in contributing to this. I get the exact same error from terraform cloud.

I have the following set up: Public Network Load Balancer with Elastic IP -> Listener on port 22 -> Target group type IP that forwards traffic to Private IP in private subnet and different availability zone

I can SSH to the instance by using my SSH key.

In terraform cloud I don't know how to get/extract the public SSH key to register in the EC2 instance.

[github-actions[bot]]

This issue is stale because it has been open 60 days with no activity. Remove stale label or comment or this will be closed in 7 days. If you don't want this issue to be closed, please set the label pinned.

[aries1980]

This is labelled as "stale" but this is still an issue.

[jsonmcfall]

This is labelled as "stale" but this is still an issue. I see the same issue, eventually after running the apply command multiple times it will work for some reason.

[github-actions[bot]]

This issue is stale because it has been open 60 days with no activity. Remove stale label or comment or this will be closed in 7 days. If you don't want this issue to be closed, please set the label pinned.

[AMKamel]

I am facing the same issue with terraform cloud as there is no way to upload and use the ssh key before initializing docker provider.

[alinefr]

This is quite an old topic as I decided to give a try to Terraform Cloud. So I had this issue about how to authenticate terraform docker provider over ssh. There's no way.

My workaround was to enable TLS over HTTPS.

I could set ca_material, cert_material & key_material as terraform variable strings sourced from Terraform Cloud variables. It's not the best thing around but it works in a good manner.