scratch_chat / chat_agent /api /performance_routes.py
WebashalarForML's picture
Upload 178 files
330b6e4 verified
"""
Performance monitoring API routes.
This module provides endpoints for monitoring system performance,
connection pool status, cache statistics, and other performance metrics.
"""
import logging
from datetime import datetime
from typing import Dict, Any
from flask import Blueprint, jsonify, current_app
from flask_limiter import Limiter
from ..utils.connection_pool import get_connection_pool_manager
from ..services.cache_service import get_cache_service
from .middleware import require_auth, create_limiter
from ..utils.response_optimization import cache_response, compress_response
logger = logging.getLogger(__name__)
# Create blueprint
performance_bp = Blueprint('performance', __name__, url_prefix='/api/v1/performance')
# Initialize rate limiter
limiter = create_limiter()
@performance_bp.route('/status', methods=['GET'])
@limiter.limit("30 per minute")
@cache_response(max_age=60, cache_type='private') # Cache for 1 minute
@compress_response
def get_performance_status():
"""
Get overall system performance status.
Returns:
JSON response with performance metrics
"""
try:
status = {
'timestamp': datetime.utcnow().isoformat(),
'status': 'healthy',
'services': {}
}
# Database connection pool status
connection_pool_manager = get_connection_pool_manager()
if connection_pool_manager:
pool_status = connection_pool_manager.get_status()
status['services']['database'] = {
'status': 'connected',
'pool_stats': pool_status['database']
}
if pool_status['redis']:
status['services']['redis'] = {
'status': 'connected' if pool_status['redis']['healthy'] else 'unhealthy',
'pool_stats': pool_status['redis']
}
else:
status['services']['redis'] = {
'status': 'disabled'
}
else:
status['services']['database'] = {'status': 'no_pool_manager'}
status['services']['redis'] = {'status': 'no_pool_manager'}
# Cache service status
cache_service = get_cache_service()
if cache_service:
cache_stats = cache_service.get_cache_stats()
status['services']['cache'] = {
'status': 'enabled' if cache_stats['redis_enabled'] else 'disabled',
'stats': cache_stats
}
else:
status['services']['cache'] = {'status': 'not_initialized'}
# Application configuration
status['configuration'] = {
'compression_enabled': current_app.config.get('ENABLE_COMPRESSION', False),
'caching_enabled': current_app.config.get('ENABLE_CACHING', False),
'performance_monitoring': current_app.config.get('ENABLE_PERFORMANCE_MONITORING', False),
'db_pool_size': current_app.config.get('DB_POOL_SIZE', 10),
'redis_max_connections': current_app.config.get('REDIS_MAX_CONNECTIONS', 20)
}
return jsonify(status)
except Exception as e:
logger.error(f"Error getting performance status: {e}")
return jsonify({
'timestamp': datetime.utcnow().isoformat(),
'status': 'error',
'error': str(e)
}), 500
@performance_bp.route('/database', methods=['GET'])
@limiter.limit("20 per minute")
@require_auth
@cache_response(max_age=30, cache_type='private')
@compress_response
def get_database_performance():
"""
Get database performance metrics.
Returns:
JSON response with database performance data
"""
try:
connection_pool_manager = get_connection_pool_manager()
if not connection_pool_manager:
return jsonify({
'error': 'Connection pool manager not available'
}), 503
# Get database pool status
pool_status = connection_pool_manager.get_status()
database_stats = pool_status['database']
# Calculate pool utilization
total_connections = database_stats['pool_size'] + database_stats['overflow']
active_connections = database_stats['checked_out']
utilization = (active_connections / total_connections * 100) if total_connections > 0 else 0
response_data = {
'timestamp': datetime.utcnow().isoformat(),
'pool_stats': database_stats,
'utilization': {
'active_connections': active_connections,
'total_connections': total_connections,
'utilization_percent': round(utilization, 2)
},
'health': {
'status': 'healthy' if utilization < 80 else 'warning' if utilization < 95 else 'critical',
'invalid_connections': database_stats['invalid']
}
}
return jsonify(response_data)
except Exception as e:
logger.error(f"Error getting database performance: {e}")
return jsonify({
'timestamp': datetime.utcnow().isoformat(),
'error': str(e)
}), 500
@performance_bp.route('/redis', methods=['GET'])
@limiter.limit("20 per minute")
@require_auth
@cache_response(max_age=30, cache_type='private')
@compress_response
def get_redis_performance():
"""
Get Redis performance metrics.
Returns:
JSON response with Redis performance data
"""
try:
connection_pool_manager = get_connection_pool_manager()
if not connection_pool_manager:
return jsonify({
'error': 'Connection pool manager not available'
}), 503
redis_client = connection_pool_manager.get_redis_client()
if not redis_client:
return jsonify({
'status': 'disabled',
'message': 'Redis is not configured'
})
# Get Redis pool status
pool_status = connection_pool_manager.get_status()
redis_stats = pool_status['redis']
# Get Redis server info
try:
redis_info = redis_client.info()
server_stats = {
'version': redis_info.get('redis_version', 'unknown'),
'uptime_seconds': redis_info.get('uptime_in_seconds', 0),
'connected_clients': redis_info.get('connected_clients', 0),
'used_memory': redis_info.get('used_memory_human', 'unknown'),
'keyspace_hits': redis_info.get('keyspace_hits', 0),
'keyspace_misses': redis_info.get('keyspace_misses', 0),
'total_commands_processed': redis_info.get('total_commands_processed', 0)
}
# Calculate hit rate
total_keyspace_ops = server_stats['keyspace_hits'] + server_stats['keyspace_misses']
hit_rate = (server_stats['keyspace_hits'] / total_keyspace_ops * 100) if total_keyspace_ops > 0 else 0
except Exception as e:
logger.warning(f"Failed to get Redis server info: {e}")
server_stats = {'error': 'Failed to get server info'}
hit_rate = 0
response_data = {
'timestamp': datetime.utcnow().isoformat(),
'pool_stats': redis_stats,
'server_stats': server_stats,
'performance': {
'hit_rate_percent': round(hit_rate, 2),
'health_status': redis_stats.get('healthy', False)
}
}
return jsonify(response_data)
except Exception as e:
logger.error(f"Error getting Redis performance: {e}")
return jsonify({
'timestamp': datetime.utcnow().isoformat(),
'error': str(e)
}), 500
@performance_bp.route('/cache', methods=['GET'])
@limiter.limit("20 per minute")
@require_auth
@cache_response(max_age=60, cache_type='private')
@compress_response
def get_cache_performance():
"""
Get cache performance metrics.
Returns:
JSON response with cache performance data
"""
try:
cache_service = get_cache_service()
if not cache_service:
return jsonify({
'status': 'not_initialized',
'message': 'Cache service not initialized'
})
cache_stats = cache_service.get_cache_stats()
# Add performance analysis
performance_analysis = {
'efficiency': 'excellent' if cache_stats['hit_rate_percent'] >= 80 else
'good' if cache_stats['hit_rate_percent'] >= 60 else
'fair' if cache_stats['hit_rate_percent'] >= 40 else 'poor',
'recommendations': []
}
if cache_stats['hit_rate_percent'] < 60:
performance_analysis['recommendations'].append('Consider increasing cache TTL')
performance_analysis['recommendations'].append('Review cache key strategies')
if cache_stats['cache_errors'] > 0:
performance_analysis['recommendations'].append('Investigate cache errors')
response_data = {
'timestamp': datetime.utcnow().isoformat(),
'cache_stats': cache_stats,
'performance_analysis': performance_analysis
}
return jsonify(response_data)
except Exception as e:
logger.error(f"Error getting cache performance: {e}")
return jsonify({
'timestamp': datetime.utcnow().isoformat(),
'error': str(e)
}), 500
@performance_bp.route('/metrics', methods=['GET'])
@limiter.limit("10 per minute")
@require_auth
@cache_response(max_age=120, cache_type='private') # Cache for 2 minutes
@compress_response
def get_comprehensive_metrics():
"""
Get comprehensive performance metrics.
Returns:
JSON response with all performance metrics
"""
try:
metrics = {
'timestamp': datetime.utcnow().isoformat(),
'system': {},
'database': {},
'redis': {},
'cache': {},
'application': {}
}
# System metrics
try:
import psutil
import os
process = psutil.Process(os.getpid())
cpu_percent = process.cpu_percent()
memory_info = process.memory_info()
metrics['system'] = {
'cpu_percent': cpu_percent,
'memory_rss_mb': round(memory_info.rss / 1024 / 1024, 2),
'memory_vms_mb': round(memory_info.vms / 1024 / 1024, 2),
'num_threads': process.num_threads(),
'num_fds': process.num_fds() if hasattr(process, 'num_fds') else 'N/A'
}
except ImportError:
metrics['system'] = {'error': 'psutil not available'}
except Exception as e:
metrics['system'] = {'error': str(e)}
# Database metrics
connection_pool_manager = get_connection_pool_manager()
if connection_pool_manager:
pool_status = connection_pool_manager.get_status()
metrics['database'] = pool_status['database']
metrics['redis'] = pool_status['redis'] or {'status': 'disabled'}
# Cache metrics
cache_service = get_cache_service()
if cache_service:
metrics['cache'] = cache_service.get_cache_stats()
# Application metrics
metrics['application'] = {
'config': {
'compression_enabled': current_app.config.get('ENABLE_COMPRESSION', False),
'caching_enabled': current_app.config.get('ENABLE_CACHING', False),
'debug_mode': current_app.config.get('DEBUG', False)
},
'flask': {
'testing': current_app.testing,
'debug': current_app.debug
}
}
return jsonify(metrics)
except Exception as e:
logger.error(f"Error getting comprehensive metrics: {e}")
return jsonify({
'timestamp': datetime.utcnow().isoformat(),
'error': str(e)
}), 500
@performance_bp.route('/health', methods=['GET'])
@limiter.limit("60 per minute")
def performance_health_check():
"""
Quick health check for performance monitoring.
Returns:
JSON response with health status
"""
try:
health_status = {
'timestamp': datetime.utcnow().isoformat(),
'status': 'healthy',
'checks': {}
}
# Database health
try:
from sqlalchemy import text
from ..models.base import db
db.session.execute(text('SELECT 1'))
health_status['checks']['database'] = 'healthy'
except Exception as e:
health_status['checks']['database'] = f'unhealthy: {str(e)}'
health_status['status'] = 'degraded'
# Redis health
connection_pool_manager = get_connection_pool_manager()
if connection_pool_manager:
redis_client = connection_pool_manager.get_redis_client()
if redis_client:
try:
redis_client.ping()
health_status['checks']['redis'] = 'healthy'
except Exception as e:
health_status['checks']['redis'] = f'unhealthy: {str(e)}'
health_status['status'] = 'degraded'
else:
health_status['checks']['redis'] = 'disabled'
else:
health_status['checks']['redis'] = 'no_pool_manager'
# Cache service health
cache_service = get_cache_service()
if cache_service:
cache_stats = cache_service.get_cache_stats()
if cache_stats['cache_errors'] > 100: # Arbitrary threshold
health_status['checks']['cache'] = 'degraded'
health_status['status'] = 'degraded'
else:
health_status['checks']['cache'] = 'healthy'
else:
health_status['checks']['cache'] = 'not_initialized'
return jsonify(health_status)
except Exception as e:
logger.error(f"Performance health check failed: {e}")
return jsonify({
'timestamp': datetime.utcnow().isoformat(),
'status': 'unhealthy',
'error': str(e)
}), 503
# Error handlers
@performance_bp.errorhandler(429)
def handle_rate_limit_exceeded(error):
"""Handle rate limit exceeded errors."""
return jsonify({
'error': 'Rate limit exceeded',
'message': 'Too many requests to performance endpoints. Please try again later.'
}), 429
@performance_bp.errorhandler(500)
def handle_internal_error(error):
"""Handle internal server errors."""
logger.error(f"Performance endpoint internal error: {error}")
return jsonify({
'error': 'Internal server error',
'timestamp': datetime.utcnow().isoformat()
}), 500