Movses AslanyanandClaude Opus 5.5
bc11638a15
Prose f-strings become past-tense events with consistent field names across services (account_id, transaction_id, amount_minor, risk_score, status_code, consumer_group, message_id, chaos_flag). The orchestrator and both consumers move from logging.getLogger to structlog so they can pass fields. New events cover what was silent before: the gateway's UpstreamFailure path (transaction_failed, with the underlying cause), fraud's score and sanctions' screening outcomes, and each successfully processed stream message. holder_name is no longer logged at all. "lookup" and "healthz" still appear in exactly one event each, so the pathology tests' log-line counts are unchanged until they move to LogQL. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
104 lines
3.9 KiB
Python
104 lines
3.9 KiB
Python
import asyncio
|
|
import json
|
|
|
|
import sqlalchemy as sa
|
|
import structlog
|
|
from opentelemetry import propagate, trace
|
|
from opentelemetry.instrumentation.utils import suppress_instrumentation
|
|
from opentelemetry.trace import SpanKind, StatusCode
|
|
|
|
from common.db import Session
|
|
from common.redis_client import redis_client
|
|
from common.settings import settings
|
|
|
|
logger = structlog.get_logger("audit")
|
|
tracer = trace.get_tracer(__name__)
|
|
|
|
GROUP = "audit"
|
|
CONSUMER = "audit-1"
|
|
|
|
|
|
async def ensure_group():
|
|
try:
|
|
await redis_client.xgroup_create(settings.stream_name, GROUP, id="$", mkstream=True)
|
|
except Exception as exc:
|
|
if "BUSYGROUP" not in str(exc):
|
|
raise
|
|
|
|
|
|
async def _handle(fields: dict):
|
|
# Store the business fields only, not the propagated trace context.
|
|
trace_fields = propagate.get_global_textmap().fields
|
|
payload = {k: v for k, v in fields.items() if k not in trace_fields}
|
|
async with Session() as session:
|
|
await session.execute(
|
|
sa.text(
|
|
"INSERT INTO audit_events (transaction_id, event_type, payload, recorded_at) "
|
|
"VALUES (:tx_id, 'transaction_processed', CAST(:payload AS jsonb), now())"
|
|
),
|
|
{"tx_id": fields.get("transaction_id"), "payload": json.dumps(payload)},
|
|
)
|
|
await session.commit()
|
|
|
|
|
|
async def run():
|
|
"""Second, independent consumer group on the same stream as `notifier` --
|
|
proves the async fork fans out to genuinely separate consumers, not one
|
|
consumer doing both jobs."""
|
|
await ensure_group()
|
|
logger.info("consumer_started", consumer_group=GROUP)
|
|
while True:
|
|
try:
|
|
# The blocking poll would otherwise be a ~2s root trace every loop.
|
|
with suppress_instrumentation():
|
|
resp = await redis_client.xreadgroup(
|
|
groupname=GROUP,
|
|
consumername=CONSUMER,
|
|
streams={settings.stream_name: ">"},
|
|
count=10,
|
|
block=2000,
|
|
)
|
|
except asyncio.CancelledError:
|
|
raise
|
|
except Exception:
|
|
logger.exception("stream_read_failed", consumer_group=GROUP)
|
|
await asyncio.sleep(1)
|
|
continue
|
|
|
|
if not resp:
|
|
continue
|
|
|
|
for _, messages in resp:
|
|
for msg_id, fields in messages:
|
|
# Parent is the gateway's PRODUCER span, carried in the message.
|
|
with tracer.start_as_current_span(
|
|
f"process {settings.stream_name}",
|
|
context=propagate.extract(fields),
|
|
kind=SpanKind.CONSUMER,
|
|
attributes={
|
|
"messaging.system": "redis",
|
|
"messaging.destination.name": settings.stream_name,
|
|
"messaging.operation.type": "process",
|
|
"messaging.consumer.group.name": GROUP,
|
|
"messaging.message.id": msg_id,
|
|
},
|
|
) as span:
|
|
try:
|
|
await _handle(fields)
|
|
logger.info(
|
|
"stream_message_processed",
|
|
consumer_group=GROUP,
|
|
message_id=msg_id,
|
|
transaction_id=fields.get("transaction_id"),
|
|
)
|
|
except Exception as exc:
|
|
span.record_exception(exc)
|
|
span.set_status(StatusCode.ERROR, str(exc))
|
|
logger.exception(
|
|
"stream_message_failed",
|
|
consumer_group=GROUP,
|
|
message_id=msg_id,
|
|
transaction_id=fields.get("transaction_id"),
|
|
)
|
|
await redis_client.xack(settings.stream_name, GROUP, msg_id)
|