Graceful Service Restart with PID File and Health Check

#!/bin/bash

SERVICE_NAME="my_web_service"
PID_FILE="/var/run/$SERVICE_NAME.pid"
LOG_FILE="/var/log/$SERVICE_NAME.log"
START_CMD="node /path/to/your/app.js > $LOG_FILE 2>&1 & echo \$! > $PID_FILE" # Example: Node.js app
HEALTH_CHECK_URL="http://localhost:3000/health" # Example: Health check endpoint
HEALTH_CHECK_RETRIES=10
HEALTH_CHECK_DELAY=2 # seconds

start_service() {
    if [ -f "$PID_FILE" ] && kill -0 "$(cat "$PID_FILE")" >/dev/null 2>&1; then
        echo "$SERVICE_NAME is already running (PID: $(cat "$PID_FILE"))."
        return 0
    fi

    echo "Starting $SERVICE_NAME..."
    eval "$START_CMD" # Use eval to execute the command with proper PID capture
    
    # Give service a moment to start and write PID
    sleep 2

    if [ -f "$PID_FILE" ] && kill -0 "$(cat "$PID_FILE")" >/dev/null 2>&1; then
        echo "$SERVICE_NAME started with PID: $(cat "$PID_FILE")."
        check_health
    else
        echo "Failed to start $SERVICE_NAME. Check $LOG_FILE for details."
        return 1
    fi
}

stop_service() {
    if [ ! -f "$PID_FILE" ]; then
        echo "$SERVICE_NAME is not running (PID file not found)."
        return 0
    fi
    
    PID=$(cat "$PID_FILE")
    if kill -0 "$PID" >/dev/null 2>&1; then
        echo "Stopping $SERVICE_NAME (PID: $PID)..."
        kill "$PID" # Send SIGTERM for graceful shutdown
        
        # Wait for the process to terminate
        for i in $(seq 1 10); do
            if ! kill -0 "$PID" >/dev/null 2>&1; then
                echo "$SERVICE_NAME stopped."
                rm -f "$PID_FILE"
                return 0
            fi
            sleep 1
        done
        echo "Warning: $SERVICE_NAME (PID: $PID) did not terminate gracefully. Forcing kill."
        kill -9 "$PID" # Force kill if not stopped gracefully
        rm -f "$PID_FILE"
        return 1
    else
        echo "$SERVICE_NAME is not running (stale PID file found). Removing $PID_FILE."
        rm -f "$PID_FILE"
        return 0
    fi
}

restart_service() {
    stop_service
    sleep 2 # Give time for port release/cleanup
    start_service
}

check_health() {
    if [ -z "$HEALTH_CHECK_URL" ]; then
        echo "No health check URL configured for $SERVICE_NAME."
        return 0
    fi
    
    echo "Performing health check for $SERVICE_NAME at $HEALTH_CHECK_URL..."
    for i in $(seq 1 $HEALTH_CHECK_RETRIES); do
        HTTP_STATUS=$(curl -s -o /dev/null -w "%{http_code}" "$HEALTH_CHECK_URL")
        if [ "$HTTP_STATUS" -eq 200 ]; then
            echo "$SERVICE_NAME health check successful (HTTP 200)."
            return 0
        else
            echo "Health check attempt $i/$HEALTH_CHECK_RETRIES: HTTP Status $HTTP_STATUS. Retrying in $HEALTH_CHECK_DELAY seconds..."
            sleep "$HEALTH_CHECK_DELAY"
        fi
    done
    echo "Error: $SERVICE_NAME health check failed after $HEALTH_CHECK_RETRIES attempts."
    return 1
}

case "$1" in
    start)
        start_service
        ;;
    stop)
        stop_service
        ;;
    restart)
        restart_service
        ;;
    status)
        if [ -f "$PID_FILE" ] && kill -0 "$(cat "$PID_FILE")" >/dev/null 2>&1; then
            echo "$SERVICE_NAME is running (PID: $(cat "$PID_FILE"))."
            check_health
        else
            echo "$SERVICE_NAME is not running."
            exit 1
        fi
        ;;
    *)
        echo "Usage: $0 {start|stop|restart|status}"
        exit 1
        ;;
esac