Add documentation

cluster/kube.tf

@@ -227,7 +227,7 @@ module "kube-hetzner" {
     },
     {
       name        = "challs-1",
-      server_type = var.scale_type,
+      server_type = var.challs_type,
       location    = var.region_1,
       labels = [
         "ressource-type=node",

cluster/tfvars/template.tfvars

@@ -35,35 +35,44 @@ cluster_dns_ctf        = "<dns-ctf-domain>"        # The domain name to use for
 # Cluster configuration
 # ------------------------
 # WARNING: Changing region while the cluster is running will cause all servers in the group to be destroyed and recreated.
-# For optimal performance, it is recommended to use the same region for all servers.
-# Region 1 is used for scale nodes and loadbalancer.
-# Possible values: fsn1, hel1, nbg1
-region_1     = "fsn1"       # Region for subgroup 1
-region_2     = "fsn1"       # Region for subgroup 2
-region_3     = "fsn1"       # Region for subgroup 3
+# For optimal performance, it is recommended to use the same region for all servers. If you want redundancy, use different regions for each group.
+# Region 1 is used for challs nodes, scale nodes and loadbalancer.
+# Possible values: fsn1, hel1, nbg1, ash, hil, sin - See https://docs.hetzner.com/cloud/general/locations/
+region_1     = "nbg1"       # Region for group 1, challs nodes, scale nodes and loadbalancer
+region_2     = "nbg1"       # Region for group 2
+region_3     = "nbg1"       # Region for group 3
 network_zone = "eu-central" # Hetzner network zone. Possible values: "eu-central", "us-east", "us-west", "ap-southeast". Regions must be within the network zone.
 
 # Servers
-# Server definitions are split into three groups: Control Plane, Agents, and Scale. Control plane and agents has three groups each, and scale has one group.
+# Server definitions are split into four groups: Control Plane, Agents, Challs and Scale. Control plane and agents has three groups each, while challs and scale is one group each.
 # Each group can be scaled and defined independently, to allow for smooth transitions between different server types and sizes.
 # Control planes are the servers that run the Kubernetes control plane, and are responsible for managing the cluster. 
 # Agents are the servers that run the workloads, and scale is used to scale the cluster up or down dynamically.
-# Scale is automatically scaled agent nodes, which is handled by the cluster autoscaler. It is optional, and can be used to scale the cluster up or down dynamically.
+# Challs are the servers that run the CTF challenges.
+# Scale is automatically scaled agent nodes, which is handled by the cluster autoscaler. It is optional, and can be used to scale the cluster up or down dynamically if there is not enough resources in the cluster.
+# Challs and scale nodes are placed in region_1, and are tainted to make normal resources prefer agent nodes, but allow scheduling on challs and scale nodes if needed.
 
 # Server types. See https://www.hetzner.com/cloud
+# Control plane nodes - Nodes that run the Kubernetes control plane components.
 control_plane_type_1 = "cx23" # Control plane group 1
 control_plane_type_2 = "cx23" # Control plane group 2
 control_plane_type_3 = "cx23" # Control plane group 3
+# Agent nodes - Nodes that run general workloads, excluding CTF challenges.
 agent_type_1         = "cx33" # Agent group 1
 agent_type_2         = "cx33" # Agent group 2
 agent_type_3         = "cx33" # Agent group 3
+# Challenge nodes - Nodes dedicated to running CTF challenges.
+challs_type          = "cx33" # CTF challenge nodes
+# Scale nodes - Nodes that are automatically scaled by the cluster autoscaler. These nodes are used to scale the cluster up or down dynamically.
 scale_type           = "cx33" # Scale group
 
 # Server count 
+# Control plane nodes - Nodes that run the Kubernetes control plane components.
 # Minimum of 1 control plane across all groups. 1 in each group is recommended for HA.
 control_plane_count_1 = 1 # Number of control plane nodes in group 1
 control_plane_count_2 = 1 # Number of control plane nodes in group 2
 control_plane_count_3 = 1 # Number of control plane nodes in group 3
+# Agent nodes - Nodes that run general workloads, excluding CTF challenges.
 # Minimum of 1 agent across all groups. 1 in each group is recommended for HA.
 agent_count_1 = 1 # Number of agent nodes in group 1
 agent_count_2 = 1 # Number of agent nodes in group 2

cluster/variables.tf

@@ -138,6 +138,11 @@ variable "agent_type_3" {
   default     = "cx32"
 }
 
+variable "challs_type" {
+  type        = string
+  description = "CTF challenge nodes server type"
+  default     = "cx32"
+}
 variable "scale_type" {
   type        = string
   description = "Scale group server type"

ctfp.py

@@ -701,12 +701,8 @@ def get_filename_tfvars(environment="test"):
         :param environment: The environment name (test, dev, prod)
         :return: The filename for the tfvars file
         '''
-
-        prefix = ""
-        if environment != "test":
-            prefix = f"{environment}."
 
-        return f"automated.{prefix}tfvars"
+        return f"automated.{environment}.tfvars"
 
     @staticmethod
     def load_tfvars(file_path: str):
@@ -895,6 +891,9 @@ def init_terraform(self, path, components: str = ""):
         try:
             # Check if tfvars file exists and is valid
             self.check_values()
+
+            # Load backend connection credentials
+            self.load_backend_credentials()
 
             # Check if backend config exists
             if not TFBackend.backend_exists(components):
@@ -1032,7 +1031,7 @@ def check_values(self):
 
         # Check if fields include "<" or ">"
         def check_placeholders(value):
-            if isinstance(value, str) and value.startswith("<") and value.endswith(">"):
+            if isinstance(value, str) and (value.startswith("<") or value.startswith("https://github.com/<")) and value.endswith(">"):
                 return True
             elif isinstance(value, dict):
                 for v in value.values():
@@ -1050,6 +1049,25 @@ def check_placeholders(value):
 
         Logger.info(f"{self.get_filename_tfvars()} is filled out correctly")
 
+
+    def load_backend_credentials(self):
+        '''
+        Load S3 backend credentials from automated.tfvars, to set Terraform S3 connection credentials
+        '''
+
+        # Load tfvars file
+        tfvars_data = TFVARS.safe_load_tfvars(self.get_path_tfvars())
+
+        # Set environment variables for S3 backend
+        os.environ["AWS_ACCESS_KEY_ID"] = tfvars_data.get("terraform_backend_s3_access_key", "")
+        os.environ["AWS_SECRET_ACCESS_KEY"] = tfvars_data.get("terraform_backend_s3_secret_key", "")
+
+        if os.environ["AWS_ACCESS_KEY_ID"] == "" or os.environ["AWS_SECRET_ACCESS_KEY"] == "":
+            Logger.error("S3 backend credentials not found in automated.tfvars. Please fill out terraform_backend_s3_access_key and terraform_backend_s3_secret_key as they are required to run the Terraform components.")
+            exit(1)
+
+        Logger.info(f"S3 backend credentials loaded")
+
     def cluster_deploy(self):
         Logger.info("Deploying the cluster")
 
@@ -1265,10 +1283,6 @@ def challenges_destroy(self):
 class CLI:
     def run(self):
         Logger.info("Starting CTF-Pilot CLI")
-        Logger.info("Checking availability of requried tools")
-        self.platform_check()
-        self.tool_check()
-        Logger.success("Required Tools are available")
 
         args = Args()
         if args.parser is None:
@@ -1298,6 +1312,11 @@ def run(self):
             args.print_help()
             exit(1)
 
+        Logger.info("Checking availability of required tools")
+        self.platform_check()
+        self.tool_check()
+        Logger.success("Required Tools are available")
+
         # Run the subcommand
         try:
             namespace.func(namespace)