Spaces:
Sleeping
Sleeping
File size: 14,175 Bytes
626b033 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 |
"""
Usage Tracking for MCP Server
Provides decorators and utilities for tracking deployment usage statistics.
Tracks request counts, response times, tool usage, and client information.
"""
import time
import functools
from typing import Optional, Callable, Any, Dict
from datetime import datetime
from sqlalchemy.orm import Session
from .database import get_db, db_transaction
from .models import UsageEvent, Deployment
# ============================================================================
# Usage Tracking Decorator
# ============================================================================
def track_usage(
deployment_id: Optional[str] = None,
tool_name: Optional[str] = None,
client_id_getter: Optional[Callable] = None,
):
"""
Decorator to track usage of MCP server functions.
Automatically records:
- Execution time
- Success/failure status
- Tool name
- Client identifier
Args:
deployment_id: Deployment ID (can be None if extracted from function args)
tool_name: Name of the tool/function being tracked
client_id_getter: Optional function to extract client ID from request
Example:
>>> @track_usage(tool_name="get_cat_facts")
>>> def get_cat_facts(deployment_id: str, count: int = 5):
>>> # Function implementation
>>> pass
>>> @track_usage(
>>> tool_name="custom_tool",
>>> client_id_getter=lambda req: req.headers.get("X-Client-ID")
>>> )
>>> def custom_tool(request, deployment_id: str):
>>> # Function implementation
>>> pass
"""
def decorator(func: Callable) -> Callable:
@functools.wraps(func)
def wrapper(*args, **kwargs):
# Extract deployment_id from arguments if not provided
dep_id = deployment_id
if dep_id is None:
# Try to get from kwargs
dep_id = kwargs.get("deployment_id")
# Try to get from first positional arg if it's a string
if dep_id is None and args and isinstance(args[0], str):
dep_id = args[0]
# Extract client_id if getter provided
client_id = None
if client_id_getter:
try:
# Try to get client_id from args/kwargs
if args:
client_id = client_id_getter(args[0])
elif kwargs:
client_id = client_id_getter(kwargs)
except Exception:
client_id = None
# Start timing
start_time = time.time()
success = True
error_msg = None
result = None
try:
# Execute the function
result = func(*args, **kwargs)
return result
except Exception as e:
success = False
error_msg = str(e)
raise
finally:
# Calculate duration
duration_ms = int((time.time() - start_time) * 1000)
# Record usage asynchronously (non-blocking)
if dep_id:
try:
record_usage_event(
deployment_id=dep_id,
tool_name=tool_name or func.__name__,
client_id=client_id,
duration_ms=duration_ms,
success=success,
error_message=error_msg,
)
except Exception as tracking_error:
# Don't let tracking errors affect the main function
print(f"Warning: Failed to record usage: {tracking_error}")
return wrapper
return decorator
# ============================================================================
# Usage Recording Functions
# ============================================================================
def record_usage_event(
deployment_id: str,
tool_name: Optional[str] = None,
client_id: Optional[str] = None,
duration_ms: Optional[int] = None,
success: bool = True,
error_message: Optional[str] = None,
metadata: Optional[Dict[str, Any]] = None,
) -> bool:
"""
Record a usage event in the database.
Args:
deployment_id: Deployment identifier
tool_name: Name of tool/function called
client_id: Client identifier
duration_ms: Request duration in milliseconds
success: Whether request succeeded
error_message: Error message if failed
metadata: Additional metadata
Returns:
bool: True if recorded successfully, False otherwise
Example:
>>> record_usage_event(
>>> deployment_id="deploy-mcp-example-123456",
>>> tool_name="get_cat_facts",
>>> duration_ms=150,
>>> success=True
>>> )
"""
try:
with db_transaction() as db:
UsageEvent.record_usage(
db=db,
deployment_id=deployment_id,
tool_name=tool_name,
client_id=client_id,
duration_ms=duration_ms,
success=success,
error_message=error_message,
metadata=metadata,
)
return True
except Exception as e:
print(f"Error recording usage event: {e}")
return False
def increment_deployment_counter(deployment_id: str, duration_ms: Optional[int] = None):
"""
Increment deployment usage counter and update statistics.
This is a lightweight alternative to recording full events.
Updates total_requests, last_used_at, and avg_response_time_ms.
Args:
deployment_id: Deployment identifier
duration_ms: Optional response time to update average
Returns:
bool: True if updated successfully, False otherwise
Example:
>>> increment_deployment_counter("deploy-mcp-example-123456", 150)
"""
try:
with db_transaction() as db:
deployment = Deployment.get_by_deployment_id(db, deployment_id)
if deployment:
if duration_ms is not None:
deployment.update_usage_stats(duration_ms)
else:
deployment.total_requests += 1
deployment.last_used_at = datetime.utcnow()
return True
except Exception as e:
print(f"Error incrementing deployment counter: {e}")
return False
# ============================================================================
# Statistics Retrieval
# ============================================================================
def get_deployment_statistics(
deployment_id: str,
days: int = 30,
) -> Optional[Dict[str, Any]]:
"""
Get usage statistics for a deployment.
Args:
deployment_id: Deployment identifier
days: Number of days to look back
Returns:
dict: Usage statistics or None if error
Example:
>>> stats = get_deployment_statistics("deploy-mcp-example-123456", days=7)
>>> print(f"Total requests: {stats['total_requests']}")
>>> print(f"Success rate: {stats['success_rate_percent']}%")
"""
try:
with get_db() as db:
stats = UsageEvent.get_stats(db, deployment_id, days)
return stats
except Exception as e:
print(f"Error getting deployment statistics: {e}")
return None
def get_tool_usage_breakdown(
deployment_id: str,
days: int = 30,
limit: int = 10,
) -> Optional[list]:
"""
Get breakdown of tool usage for a deployment.
Args:
deployment_id: Deployment identifier
days: Number of days to look back
limit: Maximum number of tools to return
Returns:
list: List of dicts with tool_name and count
Example:
>>> tools = get_tool_usage_breakdown("deploy-mcp-example-123456")
>>> for tool in tools:
>>> print(f"{tool['tool_name']}: {tool['count']} requests")
"""
try:
from sqlalchemy import and_, func
from datetime import datetime, timedelta
with get_db() as db:
cutoff_date = datetime.utcnow() - timedelta(days=days)
tool_stats = (
db.query(
UsageEvent.tool_name,
func.count(UsageEvent.id).label("count"),
)
.filter(
and_(
UsageEvent.deployment_id == deployment_id,
UsageEvent.timestamp >= cutoff_date,
UsageEvent.tool_name.isnot(None),
)
)
.group_by(UsageEvent.tool_name)
.order_by(func.count(UsageEvent.id).desc())
.limit(limit)
.all()
)
return [
{"tool_name": tool, "count": count}
for tool, count in tool_stats
]
except Exception as e:
print(f"Error getting tool usage breakdown: {e}")
return None
def get_usage_timeline(
deployment_id: str,
days: int = 7,
granularity: str = "day",
) -> Optional[list]:
"""
Get usage timeline for a deployment.
Args:
deployment_id: Deployment identifier
days: Number of days to look back
granularity: 'hour' or 'day'
Returns:
list: List of dicts with timestamp and count
Example:
>>> timeline = get_usage_timeline("deploy-mcp-example-123456", days=7)
>>> for entry in timeline:
>>> print(f"{entry['date']}: {entry['requests']} requests")
"""
try:
from sqlalchemy import and_, func
from datetime import datetime, timedelta
with get_db() as db:
cutoff_date = datetime.utcnow() - timedelta(days=days)
# Choose date truncation based on granularity
if granularity == "hour":
time_bucket = func.date_trunc("hour", UsageEvent.timestamp)
else:
time_bucket = func.date_trunc("day", UsageEvent.timestamp)
timeline_data = (
db.query(
time_bucket.label("time_bucket"),
func.count(UsageEvent.id).label("count"),
)
.filter(
and_(
UsageEvent.deployment_id == deployment_id,
UsageEvent.timestamp >= cutoff_date,
)
)
.group_by(time_bucket)
.order_by(time_bucket)
.all()
)
return [
{
"timestamp": bucket.isoformat() if bucket else None,
"requests": count,
}
for bucket, count in timeline_data
]
except Exception as e:
print(f"Error getting usage timeline: {e}")
return None
def get_client_statistics(
deployment_id: str,
days: int = 30,
limit: int = 10,
) -> Optional[list]:
"""
Get client usage statistics for a deployment.
Args:
deployment_id: Deployment identifier
days: Number of days to look back
limit: Maximum number of clients to return
Returns:
list: List of dicts with client_id and count
Example:
>>> clients = get_client_statistics("deploy-mcp-example-123456")
>>> for client in clients:
>>> print(f"Client {client['client_id']}: {client['count']} requests")
"""
try:
from sqlalchemy import and_, func
from datetime import datetime, timedelta
with get_db() as db:
cutoff_date = datetime.utcnow() - timedelta(days=days)
client_stats = (
db.query(
UsageEvent.client_id,
func.count(UsageEvent.id).label("count"),
)
.filter(
and_(
UsageEvent.deployment_id == deployment_id,
UsageEvent.timestamp >= cutoff_date,
UsageEvent.client_id.isnot(None),
)
)
.group_by(UsageEvent.client_id)
.order_by(func.count(UsageEvent.id).desc())
.limit(limit)
.all()
)
return [
{"client_id": client, "count": count}
for client, count in client_stats
]
except Exception as e:
print(f"Error getting client statistics: {e}")
return None
# ============================================================================
# Utility Functions
# ============================================================================
def get_all_deployments_stats() -> Optional[list]:
"""
Get quick statistics for all active deployments.
Returns:
list: List of dicts with deployment info and stats
Example:
>>> all_stats = get_all_deployments_stats()
>>> for deployment in all_stats:
>>> print(f"{deployment['server_name']}: {deployment['total_requests']} requests")
"""
try:
with get_db() as db:
deployments = Deployment.get_active_deployments(db)
return [
{
"deployment_id": dep.deployment_id,
"server_name": dep.server_name,
"total_requests": dep.total_requests or 0,
"last_used_at": dep.last_used_at.isoformat() if dep.last_used_at else None,
"avg_response_time_ms": dep.avg_response_time_ms,
"status": dep.status,
}
for dep in deployments
]
except Exception as e:
print(f"Error getting all deployments stats: {e}")
return None
|