Add module for Typhoon Azure with Container Linux

This commit is contained in:
Dalton Hubble 2018-08-19 18:48:22 -07:00
parent 7eb09237f4
commit 38b4ff4700
13 changed files with 1389 additions and 0 deletions

View File

@ -0,0 +1,13 @@
# Self-hosted Kubernetes assets (kubeconfig, manifests)
module "bootkube" {
source = "git::https://github.com/poseidon/terraform-render-bootkube.git?ref=70c28399703cb4ec8930394682400d90d733e5a5"
cluster_name = "${var.cluster_name}"
api_servers = ["${format("%s.%s", var.cluster_name, var.dns_zone)}"]
etcd_servers = ["${formatlist("%s.%s", azurerm_dns_a_record.etcds.*.name, var.dns_zone)}"]
asset_dir = "${var.asset_dir}"
networking = "flannel"
pod_cidr = "${var.pod_cidr}"
service_cidr = "${var.service_cidr}"
cluster_domain_suffix = "${var.cluster_domain_suffix}"
}

View File

@ -0,0 +1,163 @@
---
systemd:
units:
- name: etcd-member.service
enable: true
dropins:
- name: 40-etcd-cluster.conf
contents: |
[Service]
Environment="ETCD_IMAGE_TAG=v3.3.9"
Environment="ETCD_NAME=${etcd_name}"
Environment="ETCD_ADVERTISE_CLIENT_URLS=https://${etcd_domain}:2379"
Environment="ETCD_INITIAL_ADVERTISE_PEER_URLS=https://${etcd_domain}:2380"
Environment="ETCD_LISTEN_CLIENT_URLS=https://0.0.0.0:2379"
Environment="ETCD_LISTEN_PEER_URLS=https://0.0.0.0:2380"
Environment="ETCD_LISTEN_METRICS_URLS=http://0.0.0.0:2381"
Environment="ETCD_INITIAL_CLUSTER=${etcd_initial_cluster}"
Environment="ETCD_STRICT_RECONFIG_CHECK=true"
Environment="ETCD_SSL_DIR=/etc/ssl/etcd"
Environment="ETCD_TRUSTED_CA_FILE=/etc/ssl/certs/etcd/server-ca.crt"
Environment="ETCD_CERT_FILE=/etc/ssl/certs/etcd/server.crt"
Environment="ETCD_KEY_FILE=/etc/ssl/certs/etcd/server.key"
Environment="ETCD_CLIENT_CERT_AUTH=true"
Environment="ETCD_PEER_TRUSTED_CA_FILE=/etc/ssl/certs/etcd/peer-ca.crt"
Environment="ETCD_PEER_CERT_FILE=/etc/ssl/certs/etcd/peer.crt"
Environment="ETCD_PEER_KEY_FILE=/etc/ssl/certs/etcd/peer.key"
Environment="ETCD_PEER_CLIENT_CERT_AUTH=true"
- name: docker.service
enable: true
- name: locksmithd.service
mask: true
- name: wait-for-dns.service
enable: true
contents: |
[Unit]
Description=Wait for DNS entries
Wants=systemd-resolved.service
Before=kubelet.service
[Service]
Type=oneshot
RemainAfterExit=true
ExecStart=/bin/sh -c 'while ! /usr/bin/grep '^[^#[:space:]]' /etc/resolv.conf > /dev/null; do sleep 1; done'
[Install]
RequiredBy=kubelet.service
RequiredBy=etcd-member.service
- name: kubelet.service
enable: true
contents: |
[Unit]
Description=Kubelet via Hyperkube
Wants=rpc-statd.service
[Service]
EnvironmentFile=/etc/kubernetes/kubelet.env
Environment="RKT_RUN_ARGS=--uuid-file-save=/var/cache/kubelet-pod.uuid \
--volume=resolv,kind=host,source=/etc/resolv.conf \
--mount volume=resolv,target=/etc/resolv.conf \
--volume var-lib-cni,kind=host,source=/var/lib/cni \
--mount volume=var-lib-cni,target=/var/lib/cni \
--volume var-lib-calico,kind=host,source=/var/lib/calico \
--mount volume=var-lib-calico,target=/var/lib/calico \
--volume opt-cni-bin,kind=host,source=/opt/cni/bin \
--mount volume=opt-cni-bin,target=/opt/cni/bin \
--volume var-log,kind=host,source=/var/log \
--mount volume=var-log,target=/var/log \
--insecure-options=image"
ExecStartPre=/bin/mkdir -p /opt/cni/bin
ExecStartPre=/bin/mkdir -p /etc/kubernetes/manifests
ExecStartPre=/bin/mkdir -p /etc/kubernetes/cni/net.d
ExecStartPre=/bin/mkdir -p /etc/kubernetes/checkpoint-secrets
ExecStartPre=/bin/mkdir -p /etc/kubernetes/inactive-manifests
ExecStartPre=/bin/mkdir -p /var/lib/cni
ExecStartPre=/bin/mkdir -p /var/lib/calico
ExecStartPre=/bin/mkdir -p /var/lib/kubelet/volumeplugins
ExecStartPre=/usr/bin/bash -c "grep 'certificate-authority-data' /etc/kubernetes/kubeconfig | awk '{print $2}' | base64 -d > /etc/kubernetes/ca.crt"
ExecStartPre=-/usr/bin/rkt rm --uuid-file=/var/cache/kubelet-pod.uuid
ExecStart=/usr/lib/coreos/kubelet-wrapper \
--anonymous-auth=false \
--authentication-token-webhook \
--authorization-mode=Webhook \
--client-ca-file=/etc/kubernetes/ca.crt \
--cluster_dns=${k8s_dns_service_ip} \
--cluster_domain=${cluster_domain_suffix} \
--cni-conf-dir=/etc/kubernetes/cni/net.d \
--exit-on-lock-contention \
--kubeconfig=/etc/kubernetes/kubeconfig \
--lock-file=/var/run/lock/kubelet.lock \
--network-plugin=cni \
--node-labels=node-role.kubernetes.io/master \
--node-labels=node-role.kubernetes.io/controller="true" \
--pod-manifest-path=/etc/kubernetes/manifests \
--register-with-taints=node-role.kubernetes.io/master=:NoSchedule \
--volume-plugin-dir=/var/lib/kubelet/volumeplugins
ExecStop=-/usr/bin/rkt stop --uuid-file=/var/cache/kubelet-pod.uuid
Restart=always
RestartSec=10
[Install]
WantedBy=multi-user.target
- name: bootkube.service
contents: |
[Unit]
Description=Bootstrap a Kubernetes cluster
ConditionPathExists=!/opt/bootkube/init_bootkube.done
[Service]
Type=oneshot
RemainAfterExit=true
WorkingDirectory=/opt/bootkube
ExecStart=/opt/bootkube/bootkube-start
ExecStartPost=/bin/touch /opt/bootkube/init_bootkube.done
[Install]
WantedBy=multi-user.target
storage:
files:
- path: /etc/kubernetes/kubeconfig
filesystem: root
mode: 0644
contents:
inline: |
${kubeconfig}
- path: /etc/kubernetes/kubelet.env
filesystem: root
mode: 0644
contents:
inline: |
KUBELET_IMAGE_URL=docker://k8s.gcr.io/hyperkube
KUBELET_IMAGE_TAG=v1.11.2
- path: /etc/sysctl.d/max-user-watches.conf
filesystem: root
contents:
inline: |
fs.inotify.max_user_watches=16184
- path: /opt/bootkube/bootkube-start
filesystem: root
mode: 0544
user:
id: 500
group:
id: 500
contents:
inline: |
#!/bin/bash
# Wrapper for bootkube start
set -e
# Move experimental manifests
[ -n "$(ls /opt/bootkube/assets/manifests-*/* 2>/dev/null)" ] && mv /opt/bootkube/assets/manifests-*/* /opt/bootkube/assets/manifests && rm -rf /opt/bootkube/assets/manifests-*
BOOTKUBE_ACI="$${BOOTKUBE_ACI:-quay.io/coreos/bootkube}"
BOOTKUBE_VERSION="$${BOOTKUBE_VERSION:-v0.13.0}"
BOOTKUBE_ASSETS="$${BOOTKUBE_ASSETS:-/opt/bootkube/assets}"
exec /usr/bin/rkt run \
--trust-keys-from-https \
--volume assets,kind=host,source=$${BOOTKUBE_ASSETS} \
--mount volume=assets,target=/assets \
--volume bootstrap,kind=host,source=/etc/kubernetes \
--mount volume=bootstrap,target=/etc/kubernetes \
$${RKT_OPTS} \
$${BOOTKUBE_ACI}:$${BOOTKUBE_VERSION} \
--net=host \
--dns=host \
--exec=/bootkube -- start --asset-dir=/assets "$@"
passwd:
users:
- name: core
ssh_authorized_keys:
- "${ssh_authorized_key}"

View File

@ -0,0 +1,163 @@
# Discrete DNS records for each controller's private IPv4 for etcd usage
resource "azurerm_dns_a_record" "etcds" {
count = "${var.controller_count}"
resource_group_name = "${var.dns_zone_group}"
# DNS Zone name where record should be created
zone_name = "${var.dns_zone}"
# DNS record
name = "${format("%s-etcd%d", var.cluster_name, count.index)}"
ttl = 300
# private IPv4 address for etcd
records = ["${element(azurerm_network_interface.controllers.*.private_ip_address, count.index)}"]
}
locals {
# Channel for a Container Linux derivative
# coreos-stable -> Container Linux Stable
channel = "${element(split("-", var.os_image), 1)}"
}
# Controller availability set to spread controllers
resource "azurerm_availability_set" "controllers" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "${var.cluster_name}-controllers"
location = "${var.region}"
platform_fault_domain_count = 2
platform_update_domain_count = 4
managed = true
}
# Controller instances
resource "azurerm_virtual_machine" "controllers" {
count = "${var.controller_count}"
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "${var.cluster_name}-controller-${count.index}"
location = "${var.region}"
availability_set_id = "${azurerm_availability_set.controllers.id}"
vm_size = "${var.controller_type}"
# boot
storage_image_reference {
publisher = "CoreOS"
offer = "CoreOS"
sku = "${local.channel}"
version = "latest"
}
# storage
storage_os_disk {
name = "${var.cluster_name}-controller-${count.index}"
create_option = "FromImage"
caching = "ReadWrite"
disk_size_gb = "${var.disk_size}"
os_type = "Linux"
managed_disk_type = "Premium_LRS"
}
# network
network_interface_ids = ["${element(azurerm_network_interface.controllers.*.id, count.index)}"]
os_profile {
computer_name = "${var.cluster_name}-controller-${count.index}"
admin_username = "core"
custom_data = "${element(data.ct_config.controller-ignitions.*.rendered, count.index)}"
}
# Azure mandates setting an ssh_key, even though Ignition custom_data handles it too
os_profile_linux_config {
disable_password_authentication = true
ssh_keys {
path = "/home/core/.ssh/authorized_keys"
key_data = "${var.ssh_authorized_key}"
}
}
# lifecycle
delete_os_disk_on_termination = true
delete_data_disks_on_termination = true
lifecycle {
ignore_changes = [
"storage_os_disk",
]
}
}
# Controller NICs with public and private IPv4
resource "azurerm_network_interface" "controllers" {
count = "${var.controller_count}"
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "${var.cluster_name}-controller-${count.index}"
location = "${azurerm_resource_group.cluster.location}"
network_security_group_id = "${azurerm_network_security_group.controller.id}"
ip_configuration {
name = "ip0"
subnet_id = "${azurerm_subnet.controller.id}"
private_ip_address_allocation = "dynamic"
# public IPv4
public_ip_address_id = "${element(azurerm_public_ip.controllers.*.id, count.index)}"
# backend address pool to which the NIC should be added
load_balancer_backend_address_pools_ids = ["${azurerm_lb_backend_address_pool.controller.id}"]
}
}
# Controller public IPv4 addresses
resource "azurerm_public_ip" "controllers" {
count = "${var.controller_count}"
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "${var.cluster_name}-controller-${count.index}"
location = "${azurerm_resource_group.cluster.location}"
sku = "Standard"
public_ip_address_allocation = "static"
}
# Controller Ignition configs
data "ct_config" "controller-ignitions" {
count = "${var.controller_count}"
content = "${element(data.template_file.controller-configs.*.rendered, count.index)}"
pretty_print = false
snippets = ["${var.controller_clc_snippets}"]
}
# Controller Container Linux configs
data "template_file" "controller-configs" {
count = "${var.controller_count}"
template = "${file("${path.module}/cl/controller.yaml.tmpl")}"
vars = {
# Cannot use cyclic dependencies on controllers or their DNS records
etcd_name = "etcd${count.index}"
etcd_domain = "${var.cluster_name}-etcd${count.index}.${var.dns_zone}"
# etcd0=https://cluster-etcd0.example.com,etcd1=https://cluster-etcd1.example.com,...
etcd_initial_cluster = "${join(",", data.template_file.etcds.*.rendered)}"
kubeconfig = "${indent(10, module.bootkube.kubeconfig)}"
ssh_authorized_key = "${var.ssh_authorized_key}"
k8s_dns_service_ip = "${cidrhost(var.service_cidr, 10)}"
cluster_domain_suffix = "${var.cluster_domain_suffix}"
}
}
data "template_file" "etcds" {
count = "${var.controller_count}"
template = "etcd$${index}=https://$${cluster_name}-etcd$${index}.$${dns_zone}:2380"
vars {
index = "${count.index}"
cluster_name = "${var.cluster_name}"
dns_zone = "${var.dns_zone}"
}
}

View File

@ -0,0 +1,129 @@
# DNS record for the apiserver load balancer
resource "azurerm_dns_a_record" "apiserver" {
resource_group_name = "${var.dns_zone_group}"
# DNS Zone name where record should be created
zone_name = "${var.dns_zone}"
# DNS record
name = "${var.cluster_name}"
ttl = 300
# IPv4 address of apiserver load balancer
records = ["${azurerm_public_ip.lb-ipv4.ip_address}"]
}
# Static IPv4 address for the cluster load balancer
resource "azurerm_public_ip" "lb-ipv4" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "${var.cluster_name}-lb-ipv4"
location = "${var.region}"
sku = "Standard"
public_ip_address_allocation = "static"
}
# Network Load Balancer for apiservers and ingress
resource "azurerm_lb" "cluster" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "${var.cluster_name}"
location = "${var.region}"
sku = "Standard"
frontend_ip_configuration {
name = "public"
public_ip_address_id = "${azurerm_public_ip.lb-ipv4.id}"
}
}
resource "azurerm_lb_rule" "apiserver" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "apiserver"
loadbalancer_id = "${azurerm_lb.cluster.id}"
frontend_ip_configuration_name = "public"
protocol = "Tcp"
frontend_port = 6443
backend_port = 6443
backend_address_pool_id = "${azurerm_lb_backend_address_pool.controller.id}"
probe_id = "${azurerm_lb_probe.apiserver.id}"
}
resource "azurerm_lb_rule" "ingress-http" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "ingress-http"
loadbalancer_id = "${azurerm_lb.cluster.id}"
frontend_ip_configuration_name = "public"
protocol = "Tcp"
frontend_port = 80
backend_port = 80
backend_address_pool_id = "${azurerm_lb_backend_address_pool.worker.id}"
probe_id = "${azurerm_lb_probe.ingress.id}"
}
resource "azurerm_lb_rule" "ingress-https" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "ingress-https"
loadbalancer_id = "${azurerm_lb.cluster.id}"
frontend_ip_configuration_name = "public"
protocol = "Tcp"
frontend_port = 443
backend_port = 443
backend_address_pool_id = "${azurerm_lb_backend_address_pool.worker.id}"
probe_id = "${azurerm_lb_probe.ingress.id}"
}
# Address pool of controllers
resource "azurerm_lb_backend_address_pool" "controller" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "controller"
loadbalancer_id = "${azurerm_lb.cluster.id}"
}
# Address pool of workers
resource "azurerm_lb_backend_address_pool" "worker" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "worker"
loadbalancer_id = "${azurerm_lb.cluster.id}"
}
# Health checks / probes
# TCP health check for apiserver
resource "azurerm_lb_probe" "apiserver" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "apiserver"
loadbalancer_id = "${azurerm_lb.cluster.id}"
protocol = "Tcp"
port = 6443
# unhealthy threshold
number_of_probes = 3
interval_in_seconds = 5
}
# HTTP health check for ingress
resource "azurerm_lb_probe" "ingress" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "ingress"
loadbalancer_id = "${azurerm_lb.cluster.id}"
protocol = "Http"
port = 10254
request_path = "/healthz"
# unhealthy threshold
number_of_probes = 3
interval_in_seconds = 5
}

View File

@ -0,0 +1,33 @@
# Organize cluster into a resource group
resource "azurerm_resource_group" "cluster" {
name = "${var.cluster_name}"
location = "${var.region}"
}
resource "azurerm_virtual_network" "network" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "${var.cluster_name}"
location = "${azurerm_resource_group.cluster.location}"
address_space = ["${var.host_cidr}"]
}
# Subnets - separate subnets for controller and workers because Azure
# network security groups are based on IPv4 CIDR rather than instance
# tags like GCP or security group membership like AWS
resource "azurerm_subnet" "controller" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "controller"
virtual_network_name = "${azurerm_virtual_network.network.name}"
address_prefix = "${cidrsubnet(var.host_cidr, 1, 0)}"
}
resource "azurerm_subnet" "worker" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "worker"
virtual_network_name = "${azurerm_virtual_network.network.name}"
address_prefix = "${cidrsubnet(var.host_cidr, 1, 1)}"
}

View File

@ -0,0 +1,319 @@
# Controller security group
resource "azurerm_network_security_group" "controller" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "${var.cluster_name}-controller"
location = "${azurerm_resource_group.cluster.location}"
}
resource "azurerm_network_security_rule" "controller-ssh" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-ssh"
network_security_group_name = "${azurerm_network_security_group.controller.name}"
priority = "2000"
access = "Allow"
direction = "Inbound"
protocol = "Tcp"
source_port_range = "*"
destination_port_range = "22"
source_address_prefix = "*"
destination_address_prefix = "${azurerm_subnet.controller.address_prefix}"
}
resource "azurerm_network_security_rule" "controller-etcd" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-etcd"
network_security_group_name = "${azurerm_network_security_group.controller.name}"
priority = "2005"
access = "Allow"
direction = "Inbound"
protocol = "Tcp"
source_port_range = "*"
destination_port_range = "2379-2380"
source_address_prefix = "${azurerm_subnet.controller.address_prefix}"
destination_address_prefix = "${azurerm_subnet.controller.address_prefix}"
}
# Allow Prometheus to scrape etcd metrics
resource "azurerm_network_security_rule" "controller-etcd-metrics" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-etcd-metrics"
network_security_group_name = "${azurerm_network_security_group.controller.name}"
priority = "2010"
access = "Allow"
direction = "Inbound"
protocol = "Tcp"
source_port_range = "*"
destination_port_range = "2381"
source_address_prefix = "${azurerm_subnet.worker.address_prefix}"
destination_address_prefix = "${azurerm_subnet.controller.address_prefix}"
}
resource "azurerm_network_security_rule" "controller-apiserver" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-apiserver"
network_security_group_name = "${azurerm_network_security_group.controller.name}"
priority = "2015"
access = "Allow"
direction = "Inbound"
protocol = "Tcp"
source_port_range = "*"
destination_port_range = "6443"
source_address_prefix = "*"
destination_address_prefix = "${azurerm_subnet.controller.address_prefix}"
}
resource "azurerm_network_security_rule" "controller-flannel" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-flannel"
network_security_group_name = "${azurerm_network_security_group.controller.name}"
priority = "2020"
access = "Allow"
direction = "Inbound"
protocol = "Udp"
source_port_range = "*"
destination_port_range = "8472"
source_address_prefixes = ["${azurerm_subnet.controller.address_prefix}", "${azurerm_subnet.worker.address_prefix}"]
destination_address_prefix = "${azurerm_subnet.controller.address_prefix}"
}
# Allow Prometheus to scrape node-exporter daemonset
resource "azurerm_network_security_rule" "controller-node-exporter" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-node-exporter"
network_security_group_name = "${azurerm_network_security_group.controller.name}"
priority = "2025"
access = "Allow"
direction = "Inbound"
protocol = "Tcp"
source_port_range = "*"
destination_port_range = "9100"
source_address_prefix = "${azurerm_subnet.worker.address_prefix}"
destination_address_prefix = "${azurerm_subnet.controller.address_prefix}"
}
# Allow apiserver to access kubelet's for exec, log, port-forward
resource "azurerm_network_security_rule" "controller-kubelet" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-kubelet"
network_security_group_name = "${azurerm_network_security_group.controller.name}"
priority = "2030"
access = "Allow"
direction = "Inbound"
protocol = "Tcp"
source_port_range = "*"
destination_port_range = "10250"
# allow Prometheus to scrape kubelet metrics too
source_address_prefixes = ["${azurerm_subnet.controller.address_prefix}", "${azurerm_subnet.worker.address_prefix}"]
destination_address_prefix = "${azurerm_subnet.controller.address_prefix}"
}
# Allow heapster / metrics-server to scrape kubelet read-only
resource "azurerm_network_security_rule" "controller-kubelet-read" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-kubelet-read"
network_security_group_name = "${azurerm_network_security_group.controller.name}"
priority = "2035"
access = "Allow"
direction = "Inbound"
protocol = "Tcp"
source_port_range = "*"
destination_port_range = "10255"
source_address_prefix = "${azurerm_subnet.worker.address_prefix}"
destination_address_prefix = "${azurerm_subnet.controller.address_prefix}"
}
# Override Azure AllowVNetInBound and AllowAzureLoadBalancerInBound
# https://docs.microsoft.com/en-us/azure/virtual-network/security-overview#default-security-rules
resource "azurerm_network_security_rule" "controller-allow-loadblancer" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-loadbalancer"
network_security_group_name = "${azurerm_network_security_group.controller.name}"
priority = "3000"
access = "Allow"
direction = "Inbound"
protocol = "*"
source_port_range = "*"
destination_port_range = "*"
source_address_prefix = "AzureLoadBalancer"
destination_address_prefix = "*"
}
resource "azurerm_network_security_rule" "controller-deny-all" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "deny-all"
network_security_group_name = "${azurerm_network_security_group.controller.name}"
priority = "3005"
access = "Deny"
direction = "Inbound"
protocol = "*"
source_port_range = "*"
destination_port_range = "*"
source_address_prefix = "*"
destination_address_prefix = "*"
}
# Worker security group
resource "azurerm_network_security_group" "worker" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "${var.cluster_name}-worker"
location = "${azurerm_resource_group.cluster.location}"
}
resource "azurerm_network_security_rule" "worker-ssh" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-ssh"
network_security_group_name = "${azurerm_network_security_group.worker.name}"
priority = "2000"
access = "Allow"
direction = "Inbound"
protocol = "Tcp"
source_port_range = "*"
destination_port_range = "22"
source_address_prefix = "${azurerm_subnet.controller.address_prefix}"
destination_address_prefix = "${azurerm_subnet.worker.address_prefix}"
}
resource "azurerm_network_security_rule" "worker-http" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-http"
network_security_group_name = "${azurerm_network_security_group.worker.name}"
priority = "2005"
access = "Allow"
direction = "Inbound"
protocol = "Tcp"
source_port_range = "*"
destination_port_range = "80"
source_address_prefix = "*"
destination_address_prefix = "${azurerm_subnet.worker.address_prefix}"
}
resource "azurerm_network_security_rule" "worker-https" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-https"
network_security_group_name = "${azurerm_network_security_group.worker.name}"
priority = "2010"
access = "Allow"
direction = "Inbound"
protocol = "Tcp"
source_port_range = "*"
destination_port_range = "443"
source_address_prefix = "*"
destination_address_prefix = "${azurerm_subnet.worker.address_prefix}"
}
resource "azurerm_network_security_rule" "worker-flannel" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-flannel"
network_security_group_name = "${azurerm_network_security_group.worker.name}"
priority = "2015"
access = "Allow"
direction = "Inbound"
protocol = "Udp"
source_port_range = "*"
destination_port_range = "8472"
source_address_prefixes = ["${azurerm_subnet.controller.address_prefix}", "${azurerm_subnet.worker.address_prefix}"]
destination_address_prefix = "${azurerm_subnet.worker.address_prefix}"
}
# Allow Prometheus to scrape node-exporter daemonset
resource "azurerm_network_security_rule" "worker-node-exporter" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-node-exporter"
network_security_group_name = "${azurerm_network_security_group.worker.name}"
priority = "2020"
access = "Allow"
direction = "Inbound"
protocol = "Tcp"
source_port_range = "*"
destination_port_range = "9100"
source_address_prefix = "${azurerm_subnet.worker.address_prefix}"
destination_address_prefix = "${azurerm_subnet.worker.address_prefix}"
}
# Allow apiserver to access kubelet's for exec, log, port-forward
resource "azurerm_network_security_rule" "worker-kubelet" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-kubelet"
network_security_group_name = "${azurerm_network_security_group.worker.name}"
priority = "2025"
access = "Allow"
direction = "Inbound"
protocol = "Tcp"
source_port_range = "*"
destination_port_range = "10250"
# allow Prometheus to scrape kubelet metrics too
source_address_prefixes = ["${azurerm_subnet.controller.address_prefix}", "${azurerm_subnet.worker.address_prefix}"]
destination_address_prefix = "${azurerm_subnet.worker.address_prefix}"
}
# Allow heapster / metrics-server to scrape kubelet read-only
resource "azurerm_network_security_rule" "worker-kubelet-read" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-kubelet-read"
network_security_group_name = "${azurerm_network_security_group.worker.name}"
priority = "2030"
access = "Allow"
direction = "Inbound"
protocol = "Tcp"
source_port_range = "*"
destination_port_range = "10255"
source_address_prefix = "${azurerm_subnet.worker.address_prefix}"
destination_address_prefix = "${azurerm_subnet.worker.address_prefix}"
}
# Override Azure AllowVNetInBound and AllowAzureLoadBalancerInBound
# https://docs.microsoft.com/en-us/azure/virtual-network/security-overview#default-security-rules
resource "azurerm_network_security_rule" "worker-allow-loadblancer" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "allow-loadbalancer"
network_security_group_name = "${azurerm_network_security_group.worker.name}"
priority = "3000"
access = "Allow"
direction = "Inbound"
protocol = "*"
source_port_range = "*"
destination_port_range = "*"
source_address_prefix = "AzureLoadBalancer"
destination_address_prefix = "*"
}
resource "azurerm_network_security_rule" "worker-deny-all" {
resource_group_name = "${azurerm_resource_group.cluster.name}"
name = "deny-all"
network_security_group_name = "${azurerm_network_security_group.worker.name}"
priority = "3005"
access = "Deny"
direction = "Inbound"
protocol = "*"
source_port_range = "*"
destination_port_range = "*"
source_address_prefix = "*"
destination_address_prefix = "*"
}

View File

@ -0,0 +1,95 @@
# Secure copy etcd TLS assets to controllers.
resource "null_resource" "copy-controller-secrets" {
count = "${var.controller_count}"
depends_on = [
"azurerm_virtual_machine.controllers",
]
connection {
type = "ssh"
host = "${element(azurerm_public_ip.controllers.*.ip_address, count.index)}"
user = "core"
timeout = "15m"
}
provisioner "file" {
content = "${module.bootkube.etcd_ca_cert}"
destination = "$HOME/etcd-client-ca.crt"
}
provisioner "file" {
content = "${module.bootkube.etcd_client_cert}"
destination = "$HOME/etcd-client.crt"
}
provisioner "file" {
content = "${module.bootkube.etcd_client_key}"
destination = "$HOME/etcd-client.key"
}
provisioner "file" {
content = "${module.bootkube.etcd_server_cert}"
destination = "$HOME/etcd-server.crt"
}
provisioner "file" {
content = "${module.bootkube.etcd_server_key}"
destination = "$HOME/etcd-server.key"
}
provisioner "file" {
content = "${module.bootkube.etcd_peer_cert}"
destination = "$HOME/etcd-peer.crt"
}
provisioner "file" {
content = "${module.bootkube.etcd_peer_key}"
destination = "$HOME/etcd-peer.key"
}
provisioner "remote-exec" {
inline = [
"sudo mkdir -p /etc/ssl/etcd/etcd",
"sudo mv etcd-client* /etc/ssl/etcd/",
"sudo cp /etc/ssl/etcd/etcd-client-ca.crt /etc/ssl/etcd/etcd/server-ca.crt",
"sudo mv etcd-server.crt /etc/ssl/etcd/etcd/server.crt",
"sudo mv etcd-server.key /etc/ssl/etcd/etcd/server.key",
"sudo cp /etc/ssl/etcd/etcd-client-ca.crt /etc/ssl/etcd/etcd/peer-ca.crt",
"sudo mv etcd-peer.crt /etc/ssl/etcd/etcd/peer.crt",
"sudo mv etcd-peer.key /etc/ssl/etcd/etcd/peer.key",
"sudo chown -R etcd:etcd /etc/ssl/etcd",
"sudo chmod -R 500 /etc/ssl/etcd",
]
}
}
# Secure copy bootkube assets to ONE controller and start bootkube to perform
# one-time self-hosted cluster bootstrapping.
resource "null_resource" "bootkube-start" {
depends_on = [
"module.bootkube",
"module.workers",
"azurerm_dns_a_record.apiserver",
"null_resource.copy-controller-secrets",
]
connection {
type = "ssh"
host = "${element(azurerm_public_ip.controllers.*.ip_address, 0)}"
user = "core"
timeout = "15m"
}
provisioner "file" {
source = "${var.asset_dir}"
destination = "$HOME/assets"
}
provisioner "remote-exec" {
inline = [
"sudo mv $HOME/assets /opt/bootkube",
"sudo systemctl start bootkube",
]
}
}

View File

@ -0,0 +1,117 @@
variable "cluster_name" {
type = "string"
description = "Unique cluster name (prepended to dns_zone)"
}
# Azure
variable "region" {
type = "string"
description = "Azure Region (e.g. centralus , see `az account list-locations --output table`)"
}
variable "dns_zone" {
type = "string"
description = "Azure DNS Zone (e.g. azure.example.com)"
}
variable "dns_zone_group" {
type = "string"
description = "Resource group where the Azure DNS Zone resides (e.g. global)"
}
# instances
variable "controller_count" {
type = "string"
default = "1"
description = "Number of controllers (i.e. masters)"
}
variable "worker_count" {
type = "string"
default = "1"
description = "Number of workers"
}
variable "controller_type" {
type = "string"
default = "Standard_DS1_v2"
description = "Machine type for controllers (see `az vm list-skus --location centralus`)"
}
variable "worker_type" {
type = "string"
default = "Standard_F1"
description = "Machine type for workers (see `az vm list-skus --location centralus`)"
}
variable "os_image" {
type = "string"
default = "coreos-stable"
description = "Channel for a Container Linux derivative (coreos-stable, coreos-beta, coreos-alpha)"
}
variable "disk_size" {
type = "string"
default = "40"
description = "Size of the disk in GB"
}
variable "worker_priority" {
type = "string"
default = "Regular"
description = "Set worker priority to Low to use reduced cost surplus capacity, with the tradeoff that instances can be evicted at any time."
}
variable "controller_clc_snippets" {
type = "list"
description = "Controller Container Linux Config snippets"
default = []
}
variable "worker_clc_snippets" {
type = "list"
description = "Worker Container Linux Config snippets"
default = []
}
# configuration
variable "ssh_authorized_key" {
type = "string"
description = "SSH public key for user 'core'"
}
variable "asset_dir" {
description = "Path to a directory where generated assets should be placed (contains secrets)"
type = "string"
}
variable "host_cidr" {
description = "CIDR IPv4 range to assign to EC2 nodes"
type = "string"
default = "10.0.0.0/16"
}
variable "pod_cidr" {
description = "CIDR IPv4 range to assign Kubernetes pods"
type = "string"
default = "10.2.0.0/16"
}
variable "service_cidr" {
description = <<EOD
CIDR IPv4 range to assign Kubernetes services.
The 1st IP will be reserved for kube_apiserver, the 10th IP will be reserved for coredns.
EOD
type = "string"
default = "10.3.0.0/16"
}
variable "cluster_domain_suffix" {
description = "Queries for domains with the suffix will be answered by coredns. Default is cluster.local (e.g. foo.default.svc.cluster.local) "
type = "string"
default = "cluster.local"
}

View File

@ -0,0 +1,24 @@
module "workers" {
source = "workers"
name = "${var.cluster_name}"
# Azure
resource_group_name = "${azurerm_resource_group.cluster.name}"
region = "${azurerm_resource_group.cluster.location}"
subnet_id = "${azurerm_subnet.worker.id}"
security_group_id = "${azurerm_network_security_group.worker.id}"
backend_address_pool_id = "${azurerm_lb_backend_address_pool.worker.id}"
count = "${var.worker_count}"
vm_type = "${var.worker_type}"
os_image = "${var.os_image}"
disk_size = "${var.disk_size}"
priority = "${var.worker_priority}"
# configuration
kubeconfig = "${module.bootkube.kubeconfig}"
ssh_authorized_key = "${var.ssh_authorized_key}"
service_cidr = "${var.service_cidr}"
cluster_domain_suffix = "${var.cluster_domain_suffix}"
clc_snippets = "${var.worker_clc_snippets}"
}

View File

@ -0,0 +1,121 @@
---
systemd:
units:
- name: docker.service
enable: true
- name: locksmithd.service
mask: true
- name: wait-for-dns.service
enable: true
contents: |
[Unit]
Description=Wait for DNS entries
Wants=systemd-resolved.service
Before=kubelet.service
[Service]
Type=oneshot
RemainAfterExit=true
ExecStart=/bin/sh -c 'while ! /usr/bin/grep '^[^#[:space:]]' /etc/resolv.conf > /dev/null; do sleep 1; done'
[Install]
RequiredBy=kubelet.service
- name: kubelet.service
enable: true
contents: |
[Unit]
Description=Kubelet via Hyperkube
Wants=rpc-statd.service
[Service]
EnvironmentFile=/etc/kubernetes/kubelet.env
Environment="RKT_RUN_ARGS=--uuid-file-save=/var/cache/kubelet-pod.uuid \
--volume=resolv,kind=host,source=/etc/resolv.conf \
--mount volume=resolv,target=/etc/resolv.conf \
--volume var-lib-cni,kind=host,source=/var/lib/cni \
--mount volume=var-lib-cni,target=/var/lib/cni \
--volume var-lib-calico,kind=host,source=/var/lib/calico \
--mount volume=var-lib-calico,target=/var/lib/calico \
--volume opt-cni-bin,kind=host,source=/opt/cni/bin \
--mount volume=opt-cni-bin,target=/opt/cni/bin \
--volume var-log,kind=host,source=/var/log \
--mount volume=var-log,target=/var/log \
--insecure-options=image"
ExecStartPre=/bin/mkdir -p /opt/cni/bin
ExecStartPre=/bin/mkdir -p /etc/kubernetes/manifests
ExecStartPre=/bin/mkdir -p /etc/kubernetes/cni/net.d
ExecStartPre=/bin/mkdir -p /var/lib/cni
ExecStartPre=/bin/mkdir -p /var/lib/calico
ExecStartPre=/bin/mkdir -p /var/lib/kubelet/volumeplugins
ExecStartPre=/usr/bin/bash -c "grep 'certificate-authority-data' /etc/kubernetes/kubeconfig | awk '{print $2}' | base64 -d > /etc/kubernetes/ca.crt"
ExecStartPre=-/usr/bin/rkt rm --uuid-file=/var/cache/kubelet-pod.uuid
ExecStart=/usr/lib/coreos/kubelet-wrapper \
--anonymous-auth=false \
--authentication-token-webhook \
--authorization-mode=Webhook \
--client-ca-file=/etc/kubernetes/ca.crt \
--cluster_dns=${k8s_dns_service_ip} \
--cluster_domain=${cluster_domain_suffix} \
--cni-conf-dir=/etc/kubernetes/cni/net.d \
--exit-on-lock-contention \
--kubeconfig=/etc/kubernetes/kubeconfig \
--lock-file=/var/run/lock/kubelet.lock \
--network-plugin=cni \
--node-labels=node-role.kubernetes.io/node \
--pod-manifest-path=/etc/kubernetes/manifests \
--volume-plugin-dir=/var/lib/kubelet/volumeplugins
ExecStop=-/usr/bin/rkt stop --uuid-file=/var/cache/kubelet-pod.uuid
Restart=always
RestartSec=5
[Install]
WantedBy=multi-user.target
- name: delete-node.service
enable: true
contents: |
[Unit]
Description=Waiting to delete Kubernetes node on shutdown
[Service]
Type=oneshot
RemainAfterExit=true
ExecStart=/bin/true
ExecStop=/etc/kubernetes/delete-node
[Install]
WantedBy=multi-user.target
storage:
files:
- path: /etc/kubernetes/kubeconfig
filesystem: root
mode: 0644
contents:
inline: |
${kubeconfig}
- path: /etc/kubernetes/kubelet.env
filesystem: root
mode: 0644
contents:
inline: |
KUBELET_IMAGE_URL=docker://k8s.gcr.io/hyperkube
KUBELET_IMAGE_TAG=v1.11.2
- path: /etc/sysctl.d/max-user-watches.conf
filesystem: root
contents:
inline: |
fs.inotify.max_user_watches=16184
- path: /etc/kubernetes/delete-node
filesystem: root
mode: 0744
contents:
inline: |
#!/bin/bash
set -e
exec /usr/bin/rkt run \
--trust-keys-from-https \
--volume config,kind=host,source=/etc/kubernetes \
--mount volume=config,target=/etc/kubernetes \
--insecure-options=image \
docker://k8s.gcr.io/hyperkube:v1.11.2 \
--net=host \
--dns=host \
--exec=/kubectl -- --kubeconfig=/etc/kubernetes/kubeconfig delete node $(hostname)
passwd:
users:
- name: core
ssh_authorized_keys:
- "${ssh_authorized_key}"

View File

@ -0,0 +1 @@

View File

@ -0,0 +1,97 @@
variable "name" {
type = "string"
description = "Unique name for the worker pool"
}
# Azure
variable "region" {
type = "string"
description = "Must be set to the Azure Region of cluster"
}
variable "resource_group_name" {
type = "string"
description = "Must be set to the resource group name of cluster"
}
variable "subnet_id" {
type = "string"
description = "Must be set to the `worker_subnet_id` output by cluster"
}
variable "security_group_id" {
type = "string"
description = "Must be set to the `worker_security_group_id` output by cluster"
}
variable "backend_address_pool_id" {
type = "string"
description = "Must be set to the `worker_backend_address_pool_id` output by cluster"
}
# instances
variable "count" {
type = "string"
default = "1"
description = "Number of instances"
}
variable "vm_type" {
type = "string"
default = "Standard_DS1_v2"
description = "Machine type for instances (see `az vm list-skus --location centralus`)"
}
variable "os_image" {
type = "string"
default = "coreos-stable"
description = "Channel for a Container Linux derivative (coreos-stable, coreos-beta, coreos-alpha)"
}
variable "disk_size" {
type = "string"
default = "40"
description = "Size of the disk in GB"
}
variable "priority" {
type = "string"
default = "Regular"
description = "Set priority to Low to use reduced cost surplus capacity, with the tradeoff that instances can be evicted at any time."
}
variable "clc_snippets" {
type = "list"
description = "Container Linux Config snippets"
default = []
}
# configuration
variable "kubeconfig" {
type = "string"
description = "Must be set to `kubeconfig` output by cluster"
}
variable "ssh_authorized_key" {
type = "string"
description = "SSH public key for user 'core'"
}
variable "service_cidr" {
description = <<EOD
CIDR IPv4 range to assign Kubernetes services.
The 1st IP will be reserved for kube_apiserver, the 10th IP will be reserved for coredns.
EOD
type = "string"
default = "10.3.0.0/16"
}
variable "cluster_domain_suffix" {
description = "Queries for domains with the suffix will be answered by coredns. Default is cluster.local (e.g. foo.default.svc.cluster.local) "
type = "string"
default = "cluster.local"
}

View File

@ -0,0 +1,114 @@
locals {
# Channel for a Container Linux derivative
# coreos-stable -> Container Linux Stable
channel = "${element(split("-", var.os_image), 1)}"
}
# Workers scale set
resource "azurerm_virtual_machine_scale_set" "workers" {
resource_group_name = "${var.resource_group_name}"
name = "${var.name}-workers"
location = "${var.region}"
sku {
name = "${var.vm_type}"
tier = "standard"
capacity = "${var.count}"
}
# boot
storage_profile_image_reference {
publisher = "CoreOS"
offer = "CoreOS"
sku = "${local.channel}"
version = "latest"
}
# storage
storage_profile_os_disk {
create_option = "FromImage"
caching = "ReadWrite"
os_type = "linux"
managed_disk_type = "Standard_LRS"
}
os_profile {
computer_name_prefix = "${var.name}-worker-"
admin_username = "core"
# Required by Azure, but password auth is disabled below
admin_password = ""
custom_data = "${element(data.ct_config.worker-ignitions.*.rendered, count.index)}"
}
# Azure mandates setting an ssh_key, even though Ignition custom_data handles it too
os_profile_linux_config {
disable_password_authentication = true
ssh_keys {
path = "/home/core/.ssh/authorized_keys"
key_data = "${var.ssh_authorized_key}"
}
}
# network
network_profile {
name = "nic0"
primary = true
network_security_group_id = "${var.security_group_id}"
ip_configuration {
name = "ip0"
subnet_id = "${var.subnet_id}"
# backend address pool to which the NIC should be added
load_balancer_backend_address_pool_ids = ["${var.backend_address_pool_id}"]
}
}
# lifecycle
priority = "${var.priority}"
upgrade_policy_mode = "Manual"
}
# Scale up or down to maintain desired number, tolerating deallocations.
resource "azurerm_autoscale_setting" "workers" {
resource_group_name = "${var.resource_group_name}"
name = "maintain-desired"
location = "${var.region}"
# autoscale
enabled = true
target_resource_id = "${azurerm_virtual_machine_scale_set.workers.id}"
profile {
name = "default"
capacity {
minimum = "${var.count}"
default = "${var.count}"
maximum = "${var.count}"
}
}
}
# Worker Ignition configs
data "ct_config" "worker-ignitions" {
content = "${data.template_file.worker-configs.rendered}"
pretty_print = false
snippets = ["${var.clc_snippets}"]
}
# Worker Container Linux configs
data "template_file" "worker-configs" {
template = "${file("${path.module}/cl/worker.yaml.tmpl")}"
vars = {
kubeconfig = "${indent(10, var.kubeconfig)}"
ssh_authorized_key = "${var.ssh_authorized_key}"
k8s_dns_service_ip = "${cidrhost(var.service_cidr, 10)}"
cluster_domain_suffix = "${var.cluster_domain_suffix}"
}
}