terraform-provider-vsphere: timeout waiting for routeable interface on v0.3.0+

Terraform Version

v0.10.8

vSphere Provider Version

provider.vsphere 0.2.2 (works) provider.vsphere 0.3.0 (first version that fails) provider.vsphere 0.4.2 (also fails)

Affected Resource(s)

  • vsphere_virtual_machine

Terraform Configuration Files

provider "vsphere" {
  user                 = ""
  password             = ""
  vsphere_server       = ""
  allow_unverified_ssl = true
  version              = "0.4.2"
}

resource "vsphere_virtual_machine" "myvm" {
  name = "vm-deleteme"

  datacenter = "DC99"
  cluster    = "myclusters"

  vcpu   = "1"
  memory = "4096"

  dns_servers = ["8.8.8.8"]

  network_interface {
    label = "VM Network"
  }

  disk {
    template  = "template-20171005.1-UbuntuTrusty-DS-10-SAS-1"
    datastore = "DS-10-SAS-1"
    type      = "thin"
  }
}

Debug Output

https://gist.github.com/y0ssar1an/7c27bb3ada41d1d90814254b413a0c36

$ ps -ef | grep terraform | grep -v grep
  501 18150 16865   0 11:16AM ttys004    0:00.51 terraform apply plan
  501 18151 18150   0 11:16AM ttys004    0:00.19 /usr/local/Cellar/terraform/0.10.8/bin/terraform apply plan
  501 18152 18151   0 11:16AM ttys004    0:00.67 /Users/ryan/clusters/vsphere2/.terraform/plugins/darwin_amd64/terraform-provider-vsphere_v0.4.2_x4

Everything looks fine in vCenter.

failed_apply

Expected Behavior

The VM should create successfully and terraform should return exit code 0.

Actual Behavior

  1. timeout waiting for routeable interface.
  2. terraform returns exit code 1

Steps to Reproduce

  1. terraform plan -out plan
  2. terraform apply plan

Important Factoids

AFAIK there’s nothing unique about our vSphere configuration. We’re using Ubuntu 14.04.

About this issue

  • Original URL
  • State: closed
  • Created 7 years ago
  • Comments: 24 (8 by maintainers)

Most upvoted comments

Thanks everyone for the feedback.

We will be making some changes to the network waiter to make the routeable portion opt-out, as mentioned in #337. Keep an eye out for this in future releases.

In my testing, I could see in vCenter that the VMs were getting IP addresses. I confirmed they were reachable by ssh-ing into the VMs. But terraform was still waiting for a routeable IP address and eventually timed out. There may be a bug in the wait code.