Added wait condition until the node count equals the desired capacity.

modules/graphdb/main.tf

@@ -87,6 +87,11 @@ resource "aws_autoscaling_group" "graphdb_auto_scaling_group" {
 
   target_group_arns = var.graphdb_target_group_arns
 
+  instance_maintenance_policy {
+    min_healthy_percentage = var.instance_maintenance_policy_min_healthy_percentage
+    max_healthy_percentage = var.instance_maintenance_policy_max_healthy_percentage
+  }
+
   launch_template {
     id      = aws_launch_template.graphdb.id
     version = aws_launch_template.graphdb.latest_version
@@ -119,3 +124,4 @@ resource "aws_autoscaling_group" "graphdb_auto_scaling_group" {
     }
   }
 }
+

modules/graphdb/templates/00_functions.sh

@@ -1,12 +1,90 @@
 #!/usr/bin/env bash
 
-# Generic helper functions
+# Path to the flag file
+ASG_CHECK_FLAG="/tmp/asg_check_complete"
 
-# Function to print messages with timestamps
+# Function to log messages with a timestamp
 log_with_timestamp() {
   echo "$(date '+%Y-%m-%d %H:%M:%S'): $1"
 }
 
+# Function to check ASG node counts
+wait_for_asg_nodes() {
+  local ASG_NAME="$1"
+  local RETRY_DELAY=10
+  # Get the desired capacity of the ASG
+  local NODE_COUNT
+  NODE_COUNT=$(aws autoscaling describe-auto-scaling-groups \
+    --auto-scaling-group-names "$ASG_NAME" \
+    --query "AutoScalingGroups[0].DesiredCapacity" \
+    --output text)
+
+  if [ "$NODE_COUNT" == "None" ]; then
+    log_with_timestamp "Error: Unable to retrieve Desired Capacity for ASG: $ASG_NAME."
+    exit 1
+  fi
+
+  log_with_timestamp "Checking ASG node count for $ASG_NAME with desired node count: $NODE_COUNT"
+
+  while true; do
+    # Check InService and Terminating states via ASG
+    local IN_SERVICE_NODE_COUNT
+    IN_SERVICE_NODE_COUNT=$(aws autoscaling describe-auto-scaling-groups \
+      --auto-scaling-group-names "$ASG_NAME" \
+      --query "AutoScalingGroups[0].Instances[?LifecycleState=='InService'] | length(@)" \
+      --output text)
+
+    local TERMINATING_NODE_COUNT
+    TERMINATING_NODE_COUNT=$(aws autoscaling describe-auto-scaling-groups \
+      --auto-scaling-group-names "$ASG_NAME" \
+      --query "AutoScalingGroups[0].Instances[?LifecycleState=='Terminating'] | length(@)" \
+      --output text)
+
+    local SHUTTING_DOWN_NODE_COUNT
+    SHUTTING_DOWN_NODE_COUNT=$(aws ec2 describe-instances \
+      --filters "Name=instance-state-name,Values=shutting-down" \
+      --query "Reservations[].Instances[].InstanceId | length(@)" \
+      --output text)
+
+    log_with_timestamp "InService: $IN_SERVICE_NODE_COUNT, Terminating: $TERMINATING_NODE_COUNT, Shutting-down: $SHUTTING_DOWN_NODE_COUNT, Desired: $NODE_COUNT"
+
+    if [[ -z "$IN_SERVICE_NODE_COUNT" || "$IN_SERVICE_NODE_COUNT" -le "$NODE_COUNT" ]] \
+      && [[ "$TERMINATING_NODE_COUNT" -eq 0 ]] \
+      && [[ "$SHUTTING_DOWN_NODE_COUNT" -eq 0 ]]; then
+      log_with_timestamp "Conditions met: InService <= $NODE_COUNT, no Terminating, no Shutting-down. Proceeding..."
+      break
+    else
+      log_with_timestamp "Conditions not met. Waiting... (InService: $IN_SERVICE_NODE_COUNT, Terminating: $TERMINATING_NODE_COUNT, Shutting-down: $SHUTTING_DOWN_NODE_COUNT)"
+      sleep "$RETRY_DELAY"
+    fi
+  done
+}
+
+# Main function to wait for ASG nodes, using a flag file to ensure the check runs only once
+wait_desired_node_count() {
+  local ASG_NAME="$1"
+
+  if [ -z "$ASG_NAME" ]; then
+    log_with_timestamp "Error: ASG name not provided."
+    exit 1
+  fi
+
+  # Check if the flag file exists
+  if [ -f "$ASG_CHECK_FLAG" ]; then
+    return 0
+  fi
+
+  log_with_timestamp "Starting script execution for ASG: $ASG_NAME"
+  wait_for_asg_nodes "$ASG_NAME"
+
+  # Create the flag file to mark completion
+  touch "$ASG_CHECK_FLAG"
+  log_with_timestamp "ASG check completed. Flag created."
+}
+
+# Execute the main function with the provided ASG name
+wait_desired_node_count "${name}"
+
 # Function which waits for all DNS records to be created
 wait_dns_records() {
   local ZONE_ID="$1"

modules/graphdb/variables.tf

@@ -394,3 +394,15 @@ variable "ebs_default_kms_key" {
   description = "Define default KMS key"
   type        = string
 }
+
+variable "instance_maintenance_policy_min_healthy_percentage" {
+  description = "Define minimum healthy percentage for the Instance Maintenance Policy"
+  type        = number
+  default     = 66
+}
+
+variable "instance_maintenance_policy_max_healthy_percentage" {
+  description = "Define maximum healthy percentage for the Instance Maintenance Policy"
+  type        = number
+  default     = 100
+}