Files
unraid-mcp/unraid_mcp/subscriptions/diagnostics.py
Jacob Magar 9026faaa7c fix: correct subscription validation, byte-based log cap, partial autostart, URL scheme
- diagnostics.py: fix allow-list vs field name mismatch in subscription validator
  (_ALLOWED_SUBSCRIPTION_FIELDS now contains schema field names like "logFile",
  not operation names like "logFileSubscription", matching what _SUBSCRIPTION_NAME_PATTERN
  extracts); add _validate_subscription_query() called before any network I/O;
  replace chained .replace() URL building with build_ws_url(); gate connection_issues
  on current failure state via _analyze_subscription_status()
- manager.py: add _cap_log_content() with byte-count pre-check
  (len(value.encode("utf-8", errors="replace")) > _MAX_RESOURCE_DATA_BYTES) so
  multibyte UTF-8 content cannot bypass the 1 MB cap
- resources.py: add double-checked locking (_startup_lock) in ensure_subscriptions_started();
  propagate exception from auto_start_all_subscriptions() via raise so
  _subscriptions_started=True is never set after a failed init
- utils.py: add build_ws_url() that raises ValueError on unknown/missing URL scheme
  instead of silently falling through; add _analyze_subscription_status() helper
  that gates connection_issues on current failure state

Resolves review threads PRRT_kwDOO6Hdxs50E50Y PRRT_kwDOO6Hdxs50E50a PRRT_kwDOO6Hdxs50E50c PRRT_kwDOO6Hdxs50E50d PRRT_kwDOO6Hdxs50E2iN PRRT_kwDOO6Hdxs50E2h8
2026-03-13 10:38:17 -04:00

277 lines
11 KiB
Python

"""Subscription system troubleshooting and monitoring.
This module provides diagnostic tools for WebSocket connection testing,
subscription system monitoring, and detailed status reporting for
development and debugging purposes.
"""
import asyncio
import json
import re
from datetime import datetime
from typing import Any
import websockets
from fastmcp import FastMCP
from websockets.typing import Subprotocol
from ..config.logging import logger
from ..config.settings import UNRAID_API_KEY, UNRAID_API_URL
from ..core.exceptions import ToolError
from .manager import subscription_manager
from .resources import ensure_subscriptions_started
from .utils import _analyze_subscription_status, build_ws_ssl_context, build_ws_url
# Schema field names that appear inside the selection set of allowed subscriptions.
# The regex _SUBSCRIPTION_NAME_PATTERN extracts the first identifier after the
# opening "{", so we list the actual field names used in queries (e.g. "logFile"),
# NOT the operation-level names (e.g. "logFileSubscription").
_ALLOWED_SUBSCRIPTION_FIELDS = frozenset(
{
"logFile",
"containerStats",
"cpu",
"memory",
"array",
"network",
"docker",
"vm",
}
)
# Pattern: must start with "subscription" keyword, then extract the first selected
# field name (the word immediately after "{").
_SUBSCRIPTION_NAME_PATTERN = re.compile(r"^\s*subscription\b[^{]*\{\s*(\w+)", re.IGNORECASE)
# Reject any query that contains a bare "mutation" or "query" operation keyword.
_FORBIDDEN_KEYWORDS = re.compile(r"\b(mutation|query)\b", re.IGNORECASE)
def _validate_subscription_query(query: str) -> str:
"""Validate that a subscription query is safe to execute.
Only allows subscription operations targeting whitelisted schema field names.
Rejects any query containing mutation/query keywords.
Returns:
The extracted field name (e.g. "logFile").
Raises:
ToolError: If the query fails validation.
"""
if _FORBIDDEN_KEYWORDS.search(query):
raise ToolError("Query rejected: must be a subscription, not a mutation or query.")
match = _SUBSCRIPTION_NAME_PATTERN.match(query)
if not match:
raise ToolError(
"Query rejected: must start with 'subscription' and contain a valid "
'subscription field. Example: subscription { logFile(path: "/var/log/syslog") { content } }'
)
field_name = match.group(1)
if field_name not in _ALLOWED_SUBSCRIPTION_FIELDS:
raise ToolError(
f"Subscription field '{field_name}' is not allowed. "
f"Allowed fields: {sorted(_ALLOWED_SUBSCRIPTION_FIELDS)}"
)
return field_name
def register_diagnostic_tools(mcp: FastMCP) -> None:
"""Register diagnostic tools with the FastMCP instance.
Args:
mcp: FastMCP instance to register tools with
"""
@mcp.tool()
async def test_subscription_query(subscription_query: str) -> dict[str, Any]:
"""Test a GraphQL subscription query directly to debug schema issues.
Use this to find working subscription field names and structure.
Only whitelisted schema fields are permitted (logFile, containerStats,
cpu, memory, array, network, docker, vm).
Args:
subscription_query: The GraphQL subscription query to test
Returns:
Dict containing test results and response data
"""
# Validate before any network I/O (Bug 1 fix)
field_name = _validate_subscription_query(subscription_query)
try:
logger.info(f"[TEST_SUBSCRIPTION] Testing validated subscription field '{field_name}'")
# Build WebSocket URL — raises ValueError on invalid/missing scheme (Bug 4 fix)
try:
ws_url = build_ws_url()
except ValueError as e:
raise ToolError(str(e)) from e
ssl_context = build_ws_ssl_context(ws_url)
# Test connection
async with websockets.connect(
ws_url,
subprotocols=[Subprotocol("graphql-transport-ws"), Subprotocol("graphql-ws")],
ssl=ssl_context,
open_timeout=10,
ping_interval=30,
ping_timeout=10,
) as websocket:
# Send connection init (using standard X-API-Key format)
await websocket.send(
json.dumps(
{
"type": "connection_init",
"payload": {"headers": {"X-API-Key": UNRAID_API_KEY}},
}
)
)
# Wait for ack
response = await websocket.recv()
init_response = json.loads(response)
if init_response.get("type") != "connection_ack":
return {"error": f"Connection failed: {init_response}"}
# Send subscription
await websocket.send(
json.dumps(
{"id": "test", "type": "start", "payload": {"query": subscription_query}}
)
)
# Wait for response with timeout
try:
response = await asyncio.wait_for(websocket.recv(), timeout=5.0)
result = json.loads(response)
logger.info(f"[TEST_SUBSCRIPTION] Response: {result}")
return {"success": True, "response": result, "query_tested": subscription_query}
except TimeoutError:
return {
"success": True,
"response": "No immediate response (subscriptions may only send data on changes)",
"query_tested": subscription_query,
"note": "Connection successful, subscription may be waiting for events",
}
except ToolError:
raise
except Exception as e:
logger.error(f"[TEST_SUBSCRIPTION] Error: {e}", exc_info=True)
return {"error": str(e), "query_tested": subscription_query}
@mcp.tool()
async def diagnose_subscriptions() -> dict[str, Any]:
"""Comprehensive diagnostic tool for subscription system.
Shows detailed status, connection states, errors, and troubleshooting info.
Returns:
Dict containing comprehensive subscription system diagnostics
"""
# Ensure subscriptions are started before diagnosing
await ensure_subscriptions_started()
try:
logger.info("[DIAGNOSTIC] Running subscription diagnostics...")
# Get comprehensive status
status = subscription_manager.get_subscription_status()
# Analyze connection issues and error counts via shared helper.
# Gates connection_issues on current failure state (Bug 5 fix).
error_count, connection_issues = _analyze_subscription_status(status)
# Calculate WebSocket URL
ws_url_display: str | None = None
if UNRAID_API_URL:
try:
ws_url_display = build_ws_url()
except ValueError:
ws_url_display = None
# Add environment info with explicit typing
diagnostic_info: dict[str, Any] = {
"timestamp": datetime.now().isoformat(),
"environment": {
"auto_start_enabled": subscription_manager.auto_start_enabled,
"max_reconnect_attempts": subscription_manager.max_reconnect_attempts,
"unraid_api_url": UNRAID_API_URL[:50] + "..." if UNRAID_API_URL else None,
"api_key_configured": bool(UNRAID_API_KEY),
"websocket_url": ws_url_display,
},
"subscriptions": status,
"summary": {
"total_configured": len(subscription_manager.subscription_configs),
"auto_start_count": sum(
1
for s in subscription_manager.subscription_configs.values()
if s.get("auto_start")
),
"active_count": len(subscription_manager.active_subscriptions),
"with_data": len(subscription_manager.resource_data),
"in_error_state": error_count,
"connection_issues": connection_issues,
},
}
# Add troubleshooting recommendations
recommendations: list[str] = []
if not diagnostic_info["environment"]["api_key_configured"]:
recommendations.append(
"CRITICAL: No API key configured. Set UNRAID_API_KEY environment variable."
)
if diagnostic_info["summary"]["in_error_state"] > 0:
recommendations.append(
"Some subscriptions are in error state. Check 'connection_issues' for details."
)
if diagnostic_info["summary"]["with_data"] == 0:
recommendations.append(
"No subscriptions have received data yet. Check WebSocket connectivity and authentication."
)
if (
diagnostic_info["summary"]["active_count"]
< diagnostic_info["summary"]["auto_start_count"]
):
recommendations.append(
"Not all auto-start subscriptions are active. Check server startup logs."
)
diagnostic_info["troubleshooting"] = {
"recommendations": recommendations,
"log_commands": [
"Check server logs for [WEBSOCKET:*], [AUTH:*], [SUBSCRIPTION:*] prefixed messages",
"Look for connection timeout or authentication errors",
"Verify Unraid API URL is accessible and supports GraphQL subscriptions",
],
"next_steps": [
"If authentication fails: Verify API key has correct permissions",
"If connection fails: Check network connectivity to Unraid server",
"If no data received: Enable DEBUG logging to see detailed protocol messages",
],
}
logger.info(
f"[DIAGNOSTIC] Completed. Active: {diagnostic_info['summary']['active_count']}, With data: {diagnostic_info['summary']['with_data']}, Errors: {diagnostic_info['summary']['in_error_state']}"
)
return diagnostic_info
except Exception as e:
logger.error(f"[DIAGNOSTIC] Failed to generate diagnostics: {e}")
raise ToolError(f"Failed to generate diagnostics: {e!s}") from e
logger.info("Subscription diagnostic tools registered successfully")