Rewrite every log call as an event name with typed fields

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>
This commit is contained in:
Movses Aslanyan
2026-10-13 21:09:55 +03:30
co-authored by Claude Opus 5.5
parent 560e3afa0e
commit bc11638a15
9 changed files with 100 additions and 38 deletions
+8
View File
@@ -118,6 +118,14 @@ async def submit_transaction(body: TransactionRequest):
try:
result = await orchestrator.create_transaction(body.model_dump())
except UpstreamFailure as exc:
log = logger.error if exc.status_code >= 500 else logger.warning
log(
"transaction_failed",
account_id=body.account_id,
status_code=exc.status_code,
reason=exc.message,
cause=repr(exc.__cause__) if exc.__cause__ else None,
)
return JSONResponse(
status_code=exc.status_code, content={"error": exc.message, "request_id": request_id}
)
+17 -6
View File
@@ -1,8 +1,8 @@
import logging
import time
import uuid
import httpx
import structlog
from opentelemetry import propagate, trace
from opentelemetry.trace import SpanKind
@@ -12,7 +12,7 @@ from common.settings import settings
from . import chaos
logger = logging.getLogger("gateway")
logger = structlog.get_logger("gateway")
tracer = trace.get_tracer(__name__)
REJECT_THRESHOLD = 0.8
@@ -76,7 +76,12 @@ async def _run_chatty_health_probe(client: httpx.AsyncClient) -> None:
async def create_transaction(payload: dict) -> dict:
start = time.perf_counter()
logger.info(f"start transaction for account {payload['account_id']}")
logger.info(
"transaction_started",
account_id=payload["account_id"],
kind=payload["kind"],
amount_minor=payload["amount_minor"],
)
retry_storm = await chaos.retry_storm_enabled()
no_batching = await chaos.no_batching_enabled()
@@ -155,7 +160,7 @@ async def create_transaction(payload: dict) -> dict:
msg_id = await redis_client.xadd(settings.stream_name, fields)
span.set_attribute("messaging.message.id", msg_id)
except Exception:
logger.exception(f"failed to publish transaction {tx_id} to the stream")
logger.exception("stream_publish_failed", transaction_id=tx_id)
summary = {}
try:
@@ -165,10 +170,16 @@ async def create_transaction(payload: dict) -> dict:
if summary_resp.status_code == 200:
summary = summary_resp.json()
except httpx.RequestError:
logger.exception(f"insights summary unavailable for transaction {tx_id}")
logger.exception("insights_summary_unavailable", transaction_id=tx_id)
elapsed_ms = (time.perf_counter() - start) * 1000
logger.info(f"finished transaction {tx_id} status={settled['status']}")
logger.info(
"transaction_finished",
transaction_id=tx_id,
status=settled["status"],
risk_score=risk_score,
duration_ms=round(elapsed_ms, 1),
)
return {
"transaction_id": tx_id,