Grafana (full use of the existing monitoring/ stack, per OSS policy):
- datasources.yml: BotPostgres (uid bot-postgres) over the shared network;
grafana service joins 'shared' + carries the datasource env
- db_init.py: idempotent read-only role obs_reader (USAGE on schema public +
SELECT on bot_traces/bot_trace_events; skipped when
OBS_READER_PASSWORD empty; ALTER ROLE for rotation) + composite index
ix_bot_trace_events_stage_time (stage, created_at DESC) — senior-engineer
review applied with the column-name fix (event_type -> stage); (trace_id)
already covered by the existing (trace_id, created_at) index
- dashboards/bot-observability.json: KPI row (real traffic, events, Jev/
fallback errors, Jev p95, tokens, commits, archive guard), events/day,
per-stage p50/p95, mention statuses (arbiter+fallback aggregates + legacy
rows), DeepSeek arbiter actions, tokens/day, model-drift vs jev-1.13.0,
user-choice A/B, STT + audio.enhance health, archive size guard
- dashboards/bot-trace-explorer.json: trace table + full event drill-down
via the trace_id variable
ERP page (admin-only, Persian RTL, recharts@^3 for React 19):
- backend bot_metrics.py: read-only require_role('admin') endpoints
(summary/routing/quality/jev/stt/traces + trace detail); aggregates count
real traffic only (calibration/demo/chat_id=0 excluded)
- frontend BotMonitorPage at /bot-monitor (KPI cards, daily events,
status donut, arbiter actions, user-choice A/B, STT/enhance tables,
stage latency, tokens chart, traces table + detail drawer with payload
viewer); entry link from the admin panel; MobileTabBar hidden there
- api.ts: typed botMetrics client section
Tooling/docs:
- tools/seed_observability_demo.py: deterministic demo traces (source='demo',
every current pipeline stage incl. audio.enhance/jev.failure/arbitration/
fallback/user choices) — excluded from real-traffic aggregates
- monitoring/README.md + bot/docs/DEPLOY-AND-MONITORING.md: setup, seed and
rollback notes; alerts deliberately deferred (owner decision 2026-10-24)
Verified locally: obs_reader SELECT ok / INSERT denied; panels' SQL run as
obs_reader with data; datasource health OK; both dashboards survive a grafana
restart; endpoints 200 as admin and 403 without role; tsc + vite build clean;
page renders in-browser (admin session); bot startup provisions role+index.
155 lines
6.0 KiB
Python
155 lines
6.0 KiB
Python
"""Idempotent DDL for the bot's own tables (mirrors the api's init_db.py
|
|
convention — the ORM never creates tables; this script does, IF NOT EXISTS).
|
|
"""
|
|
import logging
|
|
|
|
from sqlalchemy import text
|
|
from sqlalchemy.ext.asyncio import AsyncSession
|
|
|
|
from app.models import Base
|
|
|
|
logger = logging.getLogger("bot.db_init")
|
|
|
|
DDL = [
|
|
"""
|
|
CREATE TABLE IF NOT EXISTS bot_links (
|
|
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
|
|
chat_id BIGINT UNIQUE NOT NULL,
|
|
user_id UUID,
|
|
casdoor_username TEXT,
|
|
display_name TEXT,
|
|
role VARCHAR(32),
|
|
created_at TIMESTAMPTZ NOT NULL DEFAULT now(),
|
|
updated_at TIMESTAMPTZ NOT NULL DEFAULT now()
|
|
)
|
|
""",
|
|
"CREATE INDEX IF NOT EXISTS ix_bot_links_chat_id ON bot_links (chat_id)",
|
|
"""
|
|
CREATE TABLE IF NOT EXISTS bot_drafts (
|
|
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
|
|
chat_id BIGINT NOT NULL,
|
|
kind VARCHAR(16) NOT NULL,
|
|
payload JSONB NOT NULL DEFAULT '{}'::jsonb,
|
|
missing JSONB,
|
|
status VARCHAR(16) NOT NULL DEFAULT 'active',
|
|
source VARCHAR(16),
|
|
source_message_id BIGINT,
|
|
review_message_id BIGINT,
|
|
created_at TIMESTAMPTZ NOT NULL DEFAULT now(),
|
|
updated_at TIMESTAMPTZ NOT NULL DEFAULT now()
|
|
)
|
|
""",
|
|
"CREATE INDEX IF NOT EXISTS ix_bot_drafts_chat_id ON bot_drafts (chat_id)",
|
|
"CREATE INDEX IF NOT EXISTS ix_bot_drafts_chat_status ON bot_drafts (chat_id, status)",
|
|
"ALTER TABLE bot_drafts ADD COLUMN IF NOT EXISTS last_bale_update_at TIMESTAMPTZ",
|
|
"ALTER TABLE bot_drafts ADD COLUMN IF NOT EXISTS source_message_id BIGINT",
|
|
"CREATE INDEX IF NOT EXISTS ix_bot_drafts_review_msg ON bot_drafts (review_message_id)",
|
|
"CREATE INDEX IF NOT EXISTS ix_bot_drafts_source_msg ON bot_drafts (source_message_id)",
|
|
"""
|
|
CREATE TABLE IF NOT EXISTS bot_messages (
|
|
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
|
|
chat_id BIGINT NOT NULL,
|
|
direction VARCHAR(4) NOT NULL,
|
|
kind VARCHAR(16) NOT NULL DEFAULT 'text',
|
|
body TEXT,
|
|
payload JSONB,
|
|
created_at TIMESTAMPTZ NOT NULL DEFAULT now()
|
|
)
|
|
""",
|
|
"CREATE INDEX IF NOT EXISTS ix_bot_messages_chat_id ON bot_messages (chat_id)",
|
|
"""
|
|
CREATE TABLE IF NOT EXISTS bot_meta (
|
|
key VARCHAR(64) PRIMARY KEY,
|
|
value TEXT,
|
|
updated_at TIMESTAMPTZ NOT NULL DEFAULT now()
|
|
)
|
|
""",
|
|
# ---- Observability (phase 1) — silent pipeline recorder, DB-only -------
|
|
"""
|
|
CREATE TABLE IF NOT EXISTS bot_traces (
|
|
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
|
|
chat_id BIGINT NOT NULL,
|
|
message_id BIGINT,
|
|
kind VARCHAR(8) NOT NULL,
|
|
source VARCHAR(16),
|
|
draft_id UUID,
|
|
draft_kind VARCHAR(16),
|
|
committed_quote_id UUID,
|
|
status VARCHAR(16) NOT NULL DEFAULT 'open',
|
|
created_at TIMESTAMPTZ NOT NULL DEFAULT now(),
|
|
updated_at TIMESTAMPTZ NOT NULL DEFAULT now()
|
|
)
|
|
""",
|
|
"CREATE INDEX IF NOT EXISTS ix_bot_traces_chat ON bot_traces (chat_id, created_at)",
|
|
"CREATE INDEX IF NOT EXISTS ix_bot_traces_draft ON bot_traces (draft_id)",
|
|
"""
|
|
CREATE TABLE IF NOT EXISTS bot_trace_events (
|
|
id BIGSERIAL PRIMARY KEY,
|
|
trace_id UUID NOT NULL REFERENCES bot_traces(id) ON DELETE CASCADE,
|
|
stage VARCHAR(32) NOT NULL,
|
|
ok BOOLEAN NOT NULL DEFAULT TRUE,
|
|
latency_ms INTEGER,
|
|
model VARCHAR(64),
|
|
field VARCHAR(32),
|
|
payload JSONB NOT NULL DEFAULT '{}'::jsonb,
|
|
created_at TIMESTAMPTZ NOT NULL DEFAULT now()
|
|
)
|
|
""",
|
|
"CREATE INDEX IF NOT EXISTS ix_bot_trace_events_trace ON bot_trace_events (trace_id, created_at)",
|
|
"CREATE INDEX IF NOT EXISTS ix_bot_trace_events_stage ON bot_trace_events (stage)",
|
|
# dashboard queries filter stage + time range together (senior-engineer review
|
|
# 2026-10-24; column is `stage` — the review's `event_type` doesn't exist here)
|
|
"CREATE INDEX IF NOT EXISTS ix_bot_trace_events_stage_time ON bot_trace_events (stage, created_at DESC)",
|
|
"CREATE INDEX IF NOT EXISTS ix_bot_trace_events_payload ON bot_trace_events USING GIN (payload jsonb_path_ops)",
|
|
]
|
|
|
|
|
|
async def ensure_obs_reader(session: AsyncSession) -> None:
|
|
"""Read-only role for the Grafana Postgres datasource (dashboard plan phase 1).
|
|
|
|
Idempotent; entirely skipped when PROCUREMENT_BOT_OBS_READER_PASSWORD is empty
|
|
(local dev may keep using the postgres superuser for the datasource instead).
|
|
Password comes from env so rotation never needs a code change.
|
|
"""
|
|
from app.config import settings
|
|
|
|
pwd = (settings.OBS_READER_PASSWORD or "").strip()
|
|
if not pwd:
|
|
logger.info("obs_reader role skipped (no OBS_READER_PASSWORD configured)")
|
|
return
|
|
# CREATE ROLE's PASSWORD clause takes a string literal only (no function
|
|
# expressions), so the env value is embedded with standard '' escaping —
|
|
# bound parameters cannot cross into a DO block.
|
|
lit = pwd.replace("'", "''")
|
|
await session.execute(text(f"""
|
|
DO $$
|
|
BEGIN
|
|
IF NOT EXISTS (SELECT 1 FROM pg_roles WHERE rolname = 'obs_reader') THEN
|
|
CREATE ROLE obs_reader LOGIN PASSWORD '{lit}';
|
|
ELSE
|
|
ALTER ROLE obs_reader WITH LOGIN PASSWORD '{lit}';
|
|
END IF;
|
|
END
|
|
$$
|
|
"""))
|
|
await session.execute(text("GRANT USAGE ON SCHEMA public TO obs_reader"))
|
|
await session.execute(text(
|
|
"GRANT SELECT ON TABLE public.bot_traces, public.bot_trace_events TO obs_reader"))
|
|
await session.commit()
|
|
logger.info("obs_reader role ready (USAGE on public + SELECT on bot_traces/bot_trace_events)")
|
|
|
|
|
|
async def ensure_bot_tables(session: AsyncSession) -> None:
|
|
for stmt in DDL:
|
|
await session.execute(text(stmt))
|
|
await session.commit()
|
|
logger.info("bot tables ready")
|
|
|
|
|
|
async def ensure_schema() -> None:
|
|
from app.db import SessionLocal
|
|
|
|
async with SessionLocal() as session:
|
|
await ensure_bot_tables(session)
|
|
await ensure_obs_reader(session)
|