Fix overly strict firewall for GCP "worker pools"

* Fix issue where worker firewall rules didn't apply to
additional workers attached to a GCP cluster using the new
"worker pools" feature (unreleased, #148). Solves host
connection timeouts and pods not being scheduled to attached
worker pools.
* Add `name` field to GCP internal worker module to represent
the unique name of of the worker pool
* Use `cluster_name` field of GCP internal worker module for
passing the name of the cluster to which workers should be
attached
This commit is contained in:
Dalton Hubble
2018-03-03 16:21:38 -08:00
parent da6aafe816
commit 45b556c08f
5 changed files with 43 additions and 31 deletions

View File

@ -79,12 +79,17 @@ Create a cluster following the Google Cloud [tutorial](../google-cloud.md#cluste
module "yavin-worker-pool" {
source = "git::https://github.com/poseidon/typhoon//google-cloud/container-linux/kubernetes/workers?ref=v1.9.4"
providers = {
google = "google.default"
}
# Google Cloud
region = "us-central1"
network = "${module.google-cloud-yavin.network_name}"
region = "us-central1"
network = "${module.google-cloud-yavin.network_name}"
cluster_name = "yavin"
# configuration
cluster_name = "yavin-16x"
name = "yavin-16x"
kubeconfig = "${module.google-cloud-yavin.kubeconfig}"
ssh_authorized_key = "${var.ssh_authorized_key}"
@ -123,7 +128,8 @@ The Google Cloud internal `workers` module supports a number of [variables](http
|:-----|:------------|:--------|
| region | Must be set to `region` of cluster | "us-central1" |
| network | Must be set to `network_name` output by cluster | "${module.cluster.network_name}" |
| cluster_name | Unique name | "yavin-worker-pool" |
| name | Unique name (distinct from cluster name) | "yavin-16x" |
| cluster_name | Must be set to `cluster_name` of cluster | "yavin" |
| kubeconfig | Must be set to `kubeconfig` output by cluster | "${module.cluster.kubeconfig}" |
| ssh_authorized_key | SSH public key for ~/.ssh_authorized_keys | "ssh-rsa AAAAB3NZ..." |