a24dae8e18
Reader + N workers architecture:
Reader: sync_poll → rate_check → dedupe → queue.put_nowait()
Workers (WORKER_CONCURRENCY, default 2): queue.get() → invoke → send → audit
Drop policy (queue full):
- put_nowait() raises QueueFull → dropped immediately (reader never blocks)
- audit matrix.queue_full + on_queue_dropped callback
- metric: matrix_bridge_queue_dropped_total{room_id,agent_id}
Graceful shutdown:
1. stop_event → reader exits loop
2. queue.join() with QUEUE_DRAIN_TIMEOUT_S (default 5s) → workers finish in-flight
3. worker tasks cancelled
New config env vars:
QUEUE_MAX_EVENTS (default 100)
WORKER_CONCURRENCY (default 2)
QUEUE_DRAIN_TIMEOUT_S (default 5)
New metrics (H3 additions):
matrix_bridge_queue_size (gauge)
matrix_bridge_queue_dropped_total (counter)
matrix_bridge_queue_wait_seconds histogram (buckets: 0.01…30s)
/health: queue.size, queue.max, queue.workers
MatrixIngressLoop: queue_size + worker_count properties
6 queue tests: enqueue/process, full-drop-audit, concurrency barrier,
graceful drain, wait metric, rate-limit-before-enqueue
Total: 71 passed
Made-with: Cursor
77 lines
2.9 KiB
Python
77 lines
2.9 KiB
Python
"""
|
|
matrix-bridge-dagi — configuration and validation
|
|
"""
|
|
import os
|
|
from dataclasses import dataclass, field
|
|
from typing import FrozenSet
|
|
|
|
|
|
@dataclass(frozen=True)
|
|
class BridgeConfig:
|
|
# Matrix homeserver
|
|
matrix_homeserver_url: str
|
|
matrix_access_token: str
|
|
matrix_user_id: str # e.g. @dagi_bridge:daarion.space
|
|
|
|
# Room → agent mapping (M1: single room)
|
|
sofiia_room_id: str # e.g. !abcdef:daarion.space
|
|
|
|
# DAGI backend
|
|
dagi_gateway_url: str # e.g. http://dagi-gateway-node1:9300
|
|
default_node_id: str # e.g. NODA1
|
|
|
|
# Sofiia Console (audit write)
|
|
sofiia_console_url: str # e.g. http://dagi-sofiia-console-node1:8002
|
|
sofiia_internal_token: str # X-Internal-Service-Token for audit ingest
|
|
|
|
# Policy
|
|
bridge_allowed_agents: FrozenSet[str]
|
|
rate_limit_room_rpm: int # max messages per room per minute
|
|
rate_limit_sender_rpm: int # max messages per sender per minute
|
|
|
|
# H2: Backpressure queue
|
|
queue_max_events: int # max pending items (drops oldest on full)
|
|
worker_concurrency: int # parallel invoke workers
|
|
queue_drain_timeout_s: float # graceful shutdown drain timeout
|
|
|
|
# Service identity
|
|
node_id: str
|
|
build_sha: str
|
|
build_time: str
|
|
|
|
|
|
def load_config() -> BridgeConfig:
|
|
"""Load and validate config from environment variables."""
|
|
|
|
def _require(key: str) -> str:
|
|
v = os.getenv(key, "").strip()
|
|
if not v:
|
|
raise RuntimeError(f"Required env var {key!r} is not set")
|
|
return v
|
|
|
|
def _optional(key: str, default: str = "") -> str:
|
|
return os.getenv(key, default).strip()
|
|
|
|
allowed_raw = _optional("BRIDGE_ALLOWED_AGENTS", "sofiia")
|
|
allowed = frozenset(a.strip() for a in allowed_raw.split(",") if a.strip())
|
|
|
|
return BridgeConfig(
|
|
matrix_homeserver_url=_require("MATRIX_HOMESERVER_URL").rstrip("/"),
|
|
matrix_access_token=_require("MATRIX_ACCESS_TOKEN"),
|
|
matrix_user_id=_require("MATRIX_USER_ID"),
|
|
sofiia_room_id=_require("SOFIIA_ROOM_ID"),
|
|
dagi_gateway_url=_require("DAGI_GATEWAY_URL").rstrip("/"),
|
|
default_node_id=_optional("DEFAULT_NODE_ID", "NODA1"),
|
|
sofiia_console_url=_optional("SOFIIA_CONSOLE_URL", "").rstrip("/"),
|
|
sofiia_internal_token=_optional("SOFIIA_INTERNAL_TOKEN", ""),
|
|
bridge_allowed_agents=allowed,
|
|
rate_limit_room_rpm=int(_optional("RATE_LIMIT_ROOM_RPM", "20")),
|
|
rate_limit_sender_rpm=int(_optional("RATE_LIMIT_SENDER_RPM", "10")),
|
|
queue_max_events=max(1, int(_optional("QUEUE_MAX_EVENTS", "100"))),
|
|
worker_concurrency=max(1, int(_optional("WORKER_CONCURRENCY", "2"))),
|
|
queue_drain_timeout_s=max(1.0, float(_optional("QUEUE_DRAIN_TIMEOUT_S", "5"))),
|
|
node_id=_optional("NODE_ID", "NODA1"),
|
|
build_sha=_optional("BUILD_SHA", "dev"),
|
|
build_time=_optional("BUILD_TIME", "local"),
|
|
)
|