* Allow multi-controller clusters on Google Cloud * GCP regional network load balancers have a long open bug in which requests originating from a backend instance are routed to the instance itself, regardless of whether the health check passes or not. As a result, only the 0th controller node registers. We've recommended just using single master GCP clusters for a while * https://issuetracker.google.com/issues/67366622 * Workaround issue by switching to a GCP TCP Proxy load balancer. TCP proxy lb routes traffic to a backend service (global) of instance group backends. In our case, spread controllers across 3 zones (all regions have 3+ zones) and organize them in 3 zonal unmanaged instance groups that serve as backends. Allows multi-controller cluster creation * GCP network load balancers only allowed legacy HTTP health checks so kubelet 10255 was checked as an approximation of controller health. Replace with TCP apiserver health checks to detect unhealth or unresponsive apiservers. * Drawbacks: GCP provision time increases, tailed logs now timeout (similar tradeoff in AWS), controllers only span 3 zones instead of the exact number in the region * Workaround in Typhoon has been known and posted for 5 months, but there still appears to be no better alternative. Its probably time to support multi-master and accept the downsides
11 KiB
Google Cloud
In this tutorial, we'll create a Kubernetes v1.10.1 cluster on Google Compute Engine (not GKE).
We'll declare a Kubernetes cluster in Terraform using the Typhoon Terraform module. On apply, a network, firewall rules, managed instance groups of Kubernetes controllers and workers, network load balancers for controllers and workers, and health checks will be created.
Controllers and workers are provisioned to run a kubelet
. A one-time bootkube bootstrap schedules an apiserver
, scheduler
, controller-manager
, and kube-dns
on controllers and runs kube-proxy
and calico
or flannel
on each node. A generated kubeconfig
provides kubectl
access to the cluster.
Requirements
- Google Cloud Account and Service Account
- Google Cloud DNS Zone (registered Domain Name or delegated subdomain)
- Terraform v0.11.x and terraform-provider-ct installed locally
Terraform Setup
Install Terraform v0.11.x on your system.
$ terraform version
Terraform v0.11.1
Add the terraform-provider-ct plugin binary for your system.
wget https://github.com/coreos/terraform-provider-ct/releases/download/v0.2.1/terraform-provider-ct-v0.2.1-linux-amd64.tar.gz
tar xzf terraform-provider-ct-v0.2.1-linux-amd64.tar.gz
sudo mv terraform-provider-ct-v0.2.1-linux-amd64/terraform-provider-ct /usr/local/bin/
Add the plugin to your ~/.terraformrc
.
providers {
ct = "/usr/local/bin/terraform-provider-ct"
}
Read concepts to learn about Terraform, modules, and organizing resources. Change to your infrastructure repository (e.g. infra
).
cd infra/clusters
Provider
Login to your Google Console API Manager and select a project, or signup if you don't have an account.
Select "Credentials", and create service account key credentials. Choose the "Compute Engine default service account" and save the JSON private key to a file that can be referenced in configs.
mv ~/Downloads/project-id-43048204.json ~/.config/google-cloud/terraform.json
Configure the Google Cloud provider to use your service account key, project-id, and region in a providers.tf
file.
provider "google" {
version = "1.6"
alias = "default"
credentials = "${file("~/.config/google-cloud/terraform.json")}"
project = "project-id"
region = "us-central1"
}
provider "local" {
version = "~> 1.0"
alias = "default"
}
provider "null" {
version = "~> 1.0"
alias = "default"
}
provider "template" {
version = "~> 1.0"
alias = "default"
}
provider "tls" {
version = "~> 1.0"
alias = "default"
}
Additional configuration options are described in the google
provider docs.
!!! tip
A project may contain multiple clusters if you wish. Regions are listed in docs or with gcloud compute regions list
.
Cluster
Define a Kubernetes cluster using the module google-cloud/container-linux/kubernetes
.
module "google-cloud-yavin" {
source = "git::https://github.com/poseidon/typhoon//google-cloud/container-linux/kubernetes?ref=v1.10.1"
providers = {
google = "google.default"
local = "local.default"
null = "null.default"
template = "template.default"
tls = "tls.default"
}
# Google Cloud
cluster_name = "yavin"
region = "us-central1"
dns_zone = "example.com"
dns_zone_name = "example-zone"
# configuration
ssh_authorized_key = "ssh-rsa AAAAB3Nz..."
asset_dir = "/home/user/.secrets/clusters/yavin"
# optional
worker_count = 2
}
Reference the variables docs or the variables.tf source.
ssh-agent
Initial bootstrapping requires bootkube.service
be started on one controller node. Terraform uses ssh-agent
to automate this step. Add your SSH private key to ssh-agent
.
ssh-add ~/.ssh/id_rsa
ssh-add -L
!!! warning
terraform apply
will hang connecting to a controller if ssh-agent
does not contain the SSH key.
Apply
Initialize the config directory if this is the first use with Terraform.
terraform init
Get or update Terraform modules.
$ terraform get # downloads missing modules
$ terraform get --update # updates all modules
Get: git::https://github.com/poseidon/typhoon (update)
Get: git::https://github.com/poseidon/bootkube-terraform.git?ref=v0.12.0 (update)
Plan the resources to be created.
$ terraform plan
Plan: 64 to add, 0 to change, 0 to destroy.
Apply the changes to create the cluster.
$ terraform apply
module.google-cloud-yavin.null_resource.bootkube-start: Still creating... (10s elapsed)
...
module.google-cloud-yavin.null_resource.bootkube-start: Still creating... (5m30s elapsed)
module.google-cloud-yavin.null_resource.bootkube-start: Still creating... (5m40s elapsed)
module.google-cloud-yavin.null_resource.bootkube-start: Creation complete (ID: 5768638456220583358)
Apply complete! Resources: 64 added, 0 changed, 0 destroyed.
In 4-8 minutes, the Kubernetes cluster will be ready.
Verify
Install kubectl on your system. Use the generated kubeconfig
credentials to access the Kubernetes cluster and list nodes.
$ export KUBECONFIG=/home/user/.secrets/clusters/yavin/auth/kubeconfig
$ kubectl get nodes
NAME STATUS AGE VERSION
yavin-controller-0.c.example-com.internal Ready 6m v1.10.1
yavin-worker-jrbf.c.example-com.internal Ready 5m v1.10.1
yavin-worker-mzdm.c.example-com.internal Ready 5m v1.10.1
List the pods.
$ kubectl get pods --all-namespaces
NAMESPACE NAME READY STATUS RESTARTS AGE
kube-system calico-node-1cs8z 2/2 Running 0 6m
kube-system calico-node-d1l5b 2/2 Running 0 6m
kube-system calico-node-sp9ps 2/2 Running 0 6m
kube-system kube-apiserver-zppls 1/1 Running 0 6m
kube-system kube-controller-manager-3271970485-gh9kt 1/1 Running 0 6m
kube-system kube-controller-manager-3271970485-h90v8 1/1 Running 1 6m
kube-system kube-dns-1187388186-zj5dl 3/3 Running 0 6m
kube-system kube-proxy-117v6 1/1 Running 0 6m
kube-system kube-proxy-9886n 1/1 Running 0 6m
kube-system kube-proxy-njn47 1/1 Running 0 6m
kube-system kube-scheduler-3895335239-5x87r 1/1 Running 0 6m
kube-system kube-scheduler-3895335239-bzrrt 1/1 Running 1 6m
kube-system pod-checkpointer-l6lrt 1/1 Running 0 6m
Going Further
Learn about maintenance and addons.
!!! note
On Container Linux clusters, install the CLUO
addon to coordinate reboots and drains when nodes auto-update. Otherwise, updates may not be applied until the next reboot.
Variables
Required
Name | Description | Example |
---|---|---|
cluster_name | Unique cluster name (prepended to dns_zone) | "yavin" |
region | Google Cloud region | "us-central1" |
dns_zone | Google Cloud DNS zone | "google-cloud.example.com" |
dns_zone_name | Google Cloud DNS zone name | "example-zone" |
ssh_authorized_key | SSH public key for user 'core' | "ssh-rsa AAAAB3NZ..." |
asset_dir | Path to a directory where generated assets should be placed (contains secrets) | "/home/user/.secrets/clusters/yavin" |
Check the list of valid regions and list Container Linux images with gcloud compute images list | grep coreos
.
DNS Zone
Clusters create a DNS A record ${cluster_name}.${dns_zone}
to resolve a network load balancer backed by controller instances. This FQDN is used by workers and kubectl
to access the apiserver. In this example, the cluster's apiserver would be accessible at yavin.google-cloud.example.com
.
You'll need a registered domain name or subdomain registered in a Google Cloud DNS zone. You can set this up once and create many clusters with unique names.
resource "google_dns_managed_zone" "zone-for-clusters" {
dns_name = "google-cloud.example.com."
name = "example-zone"
description = "Production DNS zone"
}
!!! tip "" If you have an existing domain name with a zone file elsewhere, just carve out a subdomain that can be managed on Google Cloud (e.g. google-cloud.mydomain.com) and update nameservers.
Optional
Name | Description | Default | Example |
---|---|---|---|
controller_count | Number of controllers (i.e. masters) | 1 | 3 |
worker_count | Number of workers | 1 | 3 |
controller_type | Machine type for controllers | "n1-standard-1" | See below |
worker_type | Machine type for workers | "n1-standard-1" | See below |
os_image | Container Linux image for compute instances | "coreos-stable" | "coreos-stable-1632-3-0-v20180215" |
disk_size | Size of the disk in GB | 40 | 100 |
worker_preemptible | If enabled, Compute Engine will terminate workers randomly within 24 hours | false | true |
controller_clc_snippets | Controller Container Linux Config snippets | [] | |
worker_clc_snippets | Worker Container Linux Config snippets | [] | |
networking | Choice of networking provider | "calico" | "calico" or "flannel" |
pod_cidr | CIDR IPv4 range to assign to Kubernetes pods | "10.2.0.0/16" | "10.22.0.0/16" |
service_cidr | CIDR IPv4 range to assign to Kubernetes services | "10.3.0.0/16" | "10.3.0.0/24" |
cluster_domain_suffix | FQDN suffix for Kubernetes services answered by kube-dns. | "cluster.local" | "k8s.example.com" |
Check the list of valid machine types.
Preemption
Add worker_preemeptible = "true"
to allow worker nodes to be preempted at random, but pay significantly less. Clusters tolerate stopping instances fairly well (reschedules pods, but cannot drain) and preemption provides a nice reward for running fault-tolerant cluster systems.`