Terraform STACKIT Kubernetes

Terraform module to create a K8s (SKE) Cluster with monitoring (Argus) for STACKIT cloud platform.

Consulting

You need support with STACKIT infrastructure? We are an official professional service partner of STACKIT !

Please don't hesitate to contact us at: stackit@ventx.de

---

Sie benötigen Unterstützung bei Ihrer STACKIT Infrastruktur? Wir sind offizieller Professional Service Partner von STACKIT !

Kontaktieren Sie uns via: stackit@ventx.de

Overview

This Terraform project provisions a Kubernetes cluster on STACKIT, including optional Argus monitoring and ACL controls.

Prerequisites

• Terraform v1.x.x
• STACKIT Account

Getting Started

Clone the repository: git clone https://github.com/ventx/terraform-stackit-k8s

Navigate into the directory: cd terraform-stackit-k8s

Set your Stackit Service Account Token as EnvVars:

 export STACKIT_SERVICE_ACCOUNT_EMAIL=xxx
 export STACKIT_SERVICE_ACCOUNT_TOKEN=xxx

Create terraform.tfvars to configure Node Pools (required) and other configs (optional):

# REQUIRED: ID of your STACKIT project
project_id = "12345678-1234-1234-1234-1234567890ab"

# REQUIRED: Configure Node Pool(s)
node_pools = [
  {
    name         = "np-example"
    machine_type = "c1.2"
    minimum = "2"
    maximum = "3"
    zones   = ["eu01-1", "eu01-2", "eu01-3"]
  }
]

# Enables cluster monitoring with STACKIT Argus and configure ACL to whitelist CIDRs for cluster access
extensions = {
  acl = {
    enabled       = false
    allowed_cidrs = []
  },
  argus = {
    enabled = true
  }
}

# Automatically shutdown cluster at 18:00 and start at 08:00 to save on costs
hibernations = [{
  start    = "0 8 * * *"
  end      = "0 18 * * *"
  timezone = "Europe/Berlin"
}]

Initialize Terraform: terraform init

Create Terraform plan and verify the configuration: terraform plan

Apply the Terraform configuration: terraform apply

Maintenance & Monitoring

Scheduled maintenance (auto-update of K8s and/or machine images) is controlled by the var.maintenance variable block.

Optionally enable Argus monitoring via var.monitoring variable block.

Hibernation

Auto start/stop timings for the cluster can be set through the var.hibernations variable to save on costs.

Important Notes

Terraform Providers

There are two STACKIT Terraform providers:

• Community: SchwarzIT/stackit
• Official: stackitcloud/stackit

The official provider is still buggy, so we are using the Community provider for now.

Contributing

Pull requests are very welcome. If you encounter any problems, please open an issue ❤️.

License

MIT

terraform-docs

Requirements

Name	Version
stackit	~> 1.27

Providers

Name	Version
local	n/a
stackit	~> 1.27

Modules

No modules.

Resources

Name	Type
local_sensitive_file.kubeconfig	resource
stackit_argus_instance.default	resource
stackit_kubernetes_cluster.default	resource

Inputs

Name	Description	Type	Default	Required
acl_allowed_cidrs	List of CIDRs to allow access to the cluster	list(string)	[]	no
acl_enabled	Enable / Disable ACL	bool	false	no
argus_instance	Argus instance configuration	object({ parameters = optional(map(string)) })	{ "parameters": {} }	no
argus_instance_id	Argus Instance ID for Monitoring of K8s Cluster	string	""	no
argus_plan	Name of the Argus plan to use	string	"Monitoring-Starter-EU01"	no
cluster_name	Specifies the cluster name (lower case, alphanumeric, hyphens allowed, up to 11 chars)	string	"k8s"	no
cluster_timeouts	Timeouts for cluster operations	object({ create = optional(string) update = optional(string) delete = optional(string) })	{ "create": "60m", "delete": "60m", "update": "60m" }	no
create_local_kubeconfig	Create local kubeconfig file	bool	true	no
enable_kubernetes_version_updates	Enable automatic Kubernetes version updates	bool	true	no
enable_machine_image_version_updates	Enable automatic OS image version updates	bool	true	no
extensions	Extensions configuration	object({ acl = object({ allowed_cidrs = list(string) enabled = bool }) argus = object({ enabled = bool argus_instance_id = optional(string) }) })	{ "acl": { "allowed_cidrs": [], "enabled": false }, "argus": { "enabled": false } }	no
grafana	Argus grafana configuration	object({ enable_public_access = bool })	{ "enable_public_access": false }	no
hibernations	Hibernation configuration	list(object({ start = string end = string timezone = string }))	[ { "end": "", "start": "", "timezone": "Europe/Berlin" } ]	no
k8s_version	K8s Version	string	"1.26"	no
maintenance	Maintenance window configuration	object({ enable_kubernetes_version_updates = bool enable_machine_image_version_updates = bool start = string end = string })	{ "enable_kubernetes_version_updates": true, "enable_machine_image_version_updates": true, "end": "0000-01-01T05:00:00Z", "start": "0000-01-01T03:00:00Z" }	no
maintenance_start	Start of the maintenance window	string	"02:00:00+02:00"	no
maintenance_stop	End of the maintenance window	string	"05:00:00+02:00"	no
metrics	Argus metric configuration	object({ retention_days = number retention_days_1h_downsampling = number retention_days_5m_downsampling = number })	{ "retention_days": 30, "retention_days_1h_downsampling": 3, "retention_days_5m_downsampling": 10 }	no
node_pools	Configuration for node_pools	list(object({ machine_type = string name = string container_runtime = optional(string) labels = optional(map(string)) max_surge = optional(number) max_unavailable = optional(number) maximum = optional(number) minimum = optional(number) os_name = optional(string) os_version = optional(string) taints = optional(list(object({ key = string value = string effect = string }))) volume_size_gb = optional(number) volume_type = optional(string) zones = optional(list(string)) }))	n/a	yes
project_id	stackit Project ID	any	n/a	yes

Outputs

Name	Description
argus_alerting_url	n/a
argus_dashboard_url	n/a
argus_grafana_initial_admin_password	n/a
argus_grafana_initial_admin_user	n/a
argus_grafana_url	n/a
argus_instance_id	n/a
argus_jaeger_traces_url	n/a
argus_jaeger_ui_url	n/a
argus_logs_push_url	n/a
argus_logs_url	n/a
argus_metrics_push_url	n/a
argus_metrics_url	n/a
argus_otlp_traces_url	n/a
argus_targets_url	n/a
argus_zipkin_spans_url	n/a
cluster_id	n/a
cluster_version	n/a
kube_config	n/a