#!/bin/bash

# Kind Cluster Autoscaler Simulator
# ==================================

set -e

# Configuration
CLUSTER_NAME="${KIND_CLUSTER:-docker-desktop}"
MIN_NODES=3
MAX_NODES=10
SCALE_UP_THRESHOLD=80  # CPU usage %
SCALE_DOWN_THRESHOLD=30
CHECK_INTERVAL=30

# Colors
GREEN='\033[0;32m'
YELLOW='\033[1;33m'
RED='\033[0;31m'
NC='\033[0m'

echo "🚀 Kind Cluster Autoscaler Simulator"
echo "====================================="
echo "Cluster: $CLUSTER_NAME"
echo "Min nodes: $MIN_NODES, Max nodes: $MAX_NODES"
echo ""

# Function to get current worker node count
get_node_count() {
    kubectl get nodes --no-headers | grep -v control-plane | wc -l
}

# Function to get average CPU usage
get_cpu_usage() {
    kubectl top nodes --no-headers | grep -v control-plane | \
        awk '{sum+=$3; count++} END {if(count>0) print int(sum/count); else print 0}'
}

# Function to add a node
add_node() {
    local current_count=$1
    local new_node_num=$((current_count + 1))
    local node_name="desktop-worker${new_node_num}"

    echo -e "${GREEN}📈 Scaling up: Adding node $node_name${NC}"

    # Create new Kind worker node container
    docker run -d \
        --name "$node_name" \
        --hostname "$node_name" \
        --network kind \
        --restart on-failure:1 \
        --label io.x-k8s.kind.cluster="$CLUSTER_NAME" \
        --label io.x-k8s.kind.role=worker \
        --privileged \
        --security-opt seccomp=unconfined \
        --security-opt apparmor=unconfined \
        --tmpfs /tmp \
        --tmpfs /run \
        --volume /var \
        --volume /lib/modules:/lib/modules:ro \
        kindest/node:v1.27.3

    # Wait for node to join
    sleep 10

    # Label the new node
    kubectl label node "$node_name" node-role.kubernetes.io/worker=true --overwrite

    echo -e "${GREEN}✅ Node $node_name added successfully${NC}"
}

# Function to remove a node
remove_node() {
    local node_to_remove=$(kubectl get nodes --no-headers | grep -v control-plane | tail -1 | awk '{print $1}')

    if [ -z "$node_to_remove" ]; then
        echo -e "${YELLOW}⚠️  No nodes to remove${NC}"
        return
    fi

    echo -e "${YELLOW}📉 Scaling down: Removing node $node_to_remove${NC}"

    # Drain the node
    kubectl drain "$node_to_remove" --ignore-daemonsets --delete-emptydir-data --force

    # Delete the node
    kubectl delete node "$node_to_remove"

    # Stop and remove the container
    docker stop "$node_to_remove"
    docker rm "$node_to_remove"

    echo -e "${YELLOW}✅ Node $node_to_remove removed successfully${NC}"
}

# Main monitoring loop
echo "Starting autoscaler loop (Ctrl+C to stop)..."
echo ""

while true; do
    NODE_COUNT=$(get_node_count)
    CPU_USAGE=$(get_cpu_usage)
    PENDING_PODS=$(kubectl get pods --all-namespaces --field-selector=status.phase=Pending --no-headers 2>/dev/null | wc -l)

    echo "$(date '+%H:%M:%S') - Nodes: $NODE_COUNT | CPU: ${CPU_USAGE}% | Pending Pods: $PENDING_PODS"

    # Scale up conditions
    if [ "$PENDING_PODS" -gt 0 ] || [ "$CPU_USAGE" -gt "$SCALE_UP_THRESHOLD" ]; then
        if [ "$NODE_COUNT" -lt "$MAX_NODES" ]; then
            echo -e "${GREEN}🔺 Scale up triggered (CPU: ${CPU_USAGE}%, Pending: ${PENDING_PODS})${NC}"
            add_node "$NODE_COUNT"
        else
            echo -e "${YELLOW}⚠️  Already at max nodes ($MAX_NODES)${NC}"
        fi

    # Scale down conditions
    elif [ "$CPU_USAGE" -lt "$SCALE_DOWN_THRESHOLD" ] && [ "$NODE_COUNT" -gt "$MIN_NODES" ]; then
        echo -e "${YELLOW}🔻 Scale down triggered (CPU: ${CPU_USAGE}%)${NC}"
        remove_node
    fi

    sleep "$CHECK_INTERVAL"
done